32 lines
61 KiB
JSON
32 lines
61 KiB
JSON
{"id": "fbb44", "status": "applied", "created": "2026-09-01", "pattern": "answer-self-config-from-guesswork", "severity": "high", "diagnosis": "Otázky o vlastní konfiguraci, schopnostech a volbě modelu nanobotu jsou odpovídány typovanými fakty místo ověření v dokumentaci, přestože SOUL.md nařizuje nejdřív web_fetch na nanobot.wiki docs. Vrcholí to úpravou vlastního config.json podle vymyšleného schématu, která se neprojevila; agent pak bez diagnózy opět požádal uživatele o restart.", "evidence": [{"session": "websocket_bfc38e05", "when": "2026-05-26", "excerpt": "u: byl by pro ucely nanobot vhodnejsi jiny qwen model? — doporučení qwen-plus a qwen-turbo s tvrzeními o latenci a nákladech, žádný dohledání ani docs fetch"}, {"session": "websocket_b32cc526", "when": "2026-05-26", "excerpt": "u: pouzivas pri spousteni python kodu uv tool? — odpověď Ano, odkaz na TOOLS.md, ačkoliv tentýž den dvě sessiony pouštěly python3 -c"}, {"session": "websocket_36af5714", "when": "2026-05-27", "excerpt": "tvrdé tvrzení, že vlastní slash příkazy nejsou možné a že vestavěné jsou jen /dream-log a /help, z hlavy, před jakýmkoli researchem"}, {"session": "websocket_1ae70a88", "when": "2026-05-27", "excerpt": "vymyšlené schéma tools.my.allow_set v config.json, edit_file na živý config, po restartu stále disabled — místo diagnózy opět požadavek na restart"}], "occurrences": 8, "sessions_affected": 7, "proposal": "Při každém dotazu nebo akci týkající se konfigurace a schopností nanobotu nejprve načíst oficiální dokumentaci; config.json neupravovat podle paměti, jen po ověření schématu v dokumentaci.", "patch": {"file": "SOUL.md", "old_text": "- Pro dotazy o vlastním fungování nanobot (konfigurace, schopnosti, jak funguje) → nejdřív `web_fetch` na https://nanobot.wiki/docs/0.2.0/", "new_text": "- Pro dotazy o vlastním fungování nanobot (konfigurace, schopnosti, modely, jak funguje) → **před odpovědí i před jakoukoli úpravou config.json** `web_fetch` na https://nanobot.wiki/docs/0.2.0/ — bez ověřené dokumentace netipuj fakta a neupravuj konfiguraci"}, "history": ["2026-09-01:f5092"], "applied": {"at": "2026-09-01 07:15", "sha": "26f30eb", "file": "SOUL.md"}}
|
||
{"id": "f09a7", "status": "rejected", "created": "2026-09-01", "pattern": "reimplement-without-checking-existing", "severity": "high", "diagnosis": "Architektura jednoho cron jobu + skriptu + YAML úložiště pro /remind byla implementována v jedné session (~13:00), a odpoledne téhož dne ji jiná session navrhla a implementovala znovu od nuly pod jiným názvem — vznikly dva paralelní skripty, dva cron joby a dvakrát přepsané úložiště.", "evidence": [{"session": "websocket_36af5714", "when": "2026-05-27 ~13:00", "excerpt": "implementace remind_check.py + remind-check cron job + převod reminder.md na YAML, complete_goal ohlašuje hotovo"}, {"session": "websocket_a819aa9d", "when": "2026-05-27 14:39", "excerpt": "stejný den stejný návrh single-runner architektury jako novinka: remind_runner.py + remind-runner job, smazání 8 jobů, další přepsání reminder.md; při re-read SKILL.md agent konstatuje, že soubor je úplně jiný, než čeká, a pokračuje dál bez zjištění proč"}], "occurrences": 4, "sessions_affected": 5, "proposal": "Před návrhem nebo implementací architektury vždy nejdřív vypsat aktuální stav workspace pro danou feature (scripts/, cron list, SKILL.md) a zjistit, zda už podobná implementace neexistuje.", "history": ["2026-09-01:f1b4c"], "rejected_at": "2026-09-01 10:55"}
|
||
{"id": "f4f66", "status": "applied", "created": "2026-09-01", "pattern": "unverified-success-claim", "severity": "high", "diagnosis": "Úspěch/uložení je ohlašován bez důkazu — včetně situací těsně po selhání zápisu, po slibu opravy bez jediného tool calu a po prezentace obsahu souboru bez úspěšného načtení.", "evidence": [{"session": "websocket_9fc24346", "when": "2026-05-27 07:52", "excerpt": "my(set) → ERROR set is disabled; přesto odpověď jen Poznamenáno. bez upozornění, že se nic neuložilo"}, {"session": "websocket_b5891423", "when": "2026-05-27 07:36", "excerpt": "uživatel upozorní na chybějící diakritiku, agent slíbí opravu, neprovede žádný tool call, pak vypíše tři úkoly s diakritikou, ačkoliv read_file call selhal (leaked) a soubor obsahoval dva úkoly bez diakritiky"}, {"session": "websocket_36af5714", "when": "2026-05-27", "excerpt": "agent ohlásí Tady je celý obsah skills/remind/SKILL.md, uživatel odpovídá nic nevidim — obsah nebyl doručen, musel se posílat znovu"}], "occurrences": 8, "sessions_affected": 8, "proposal": "Nikdy neohlašovat uložení ani splnění bez úspěšného výsledku toolu; po selhání nebo leaku call selhání přiznat a operaci provést znovu.", "patch": {"file": "SOUL.md", "old_text": "- Po vícekrokových změnách ověř výsledek (re-read, test, kontrola výstupu)", "new_text": "- **Neohlašuj splnění/uložení bez úspěšného výsledku toolu** — po selhání nebo leaku callu selhání přiznej a operaci zopakuj\n- Po vícekrokových změnách ověř výsledek (re-read, test, kontrola výstupu)"}, "history": ["2026-09-01:f9d0d"], "applied": {"at": "2026-09-01 07:13", "sha": "b34d23f", "file": "SOUL.md"}}
|
||
{"id": "ff77b", "status": "applied", "created": "2026-09-01", "pattern": "retry-without-diagnosis", "severity": "medium", "diagnosis": "Po selhání nebo zkráceném výsledku web_fetch se stejný call opakuje se stejnými nebo poškozenými argumenty místo diagnózy režimu selhání (bot ochrana, paywall, guard proti opakovaným lookupům).", "evidence": [{"session": "websocket_2c3e3ae3", "when": "2026-05-27", "excerpt": "~8 web_fetch pokusů na týž Medium článek; výstupy 455–933 B zjevně indikují blokaci; jeden call má poškozenou dvojitou r.jina.ai URL (r.jina.ai/http://r.jina.ai/http://medium.com/…) a stejný cíl se fetchuje znovu i po zjištění bot ochrany"}, {"session": "websocket_36af5714", "when": "2026-05-27", "excerpt": "web_fetch na discussions/431 dvakrát zablokován guardem repeated external lookup blocked, následují další téměř identické search variace; předtím tři cat+python parse pokusy tool-result souboru, dvrátí 13 B, až potřetí se argument opraví"}], "occurrences": 18, "sessions_affected": 4, "proposal": "Po dvou po sobě jdoucích neúspěšných nebo výrazně zkrácených fetších téhož cíle zastavit, pojmenovat pozorovaný režim selhání a změnit strategii (jiný zdroj, požádat uživatele o text), nikoli opakovat tentýž call.", "patch": {"file": "SOUL.md", "old_text": "- Když tool call selže, diagnostikuj a zkus jiný přístup, než ohlásíš neúspěch", "new_text": "- Když tool call selže, diagnostikuj a zkus jiný přístup, než ohlásíš neúspěch\n- Po dvou neúspěšných nebo výrazně zkrácených fetších téhož cíle **zastav a pojmenuj režim selhání** (bot ochrana, paywall, guard) — změň strategii, neopakuj tentýž call"}, "history": ["2026-09-01:f70dc"], "applied": {"at": "2026-09-01 07:17", "sha": "540c599", "file": "SOUL.md"}}
|
||
{"id": "ff93a", "status": "rejected", "created": "2026-09-01", "pattern": "correction-not-applied", "severity": "medium", "diagnosis": "Poté, co uživatel opravil pravidlo pro zápis poznámek (psát s diakritikou), agent opravu potvrdil, ale stávající záznamy nikdy neopravil a následné zápisy ve třech dalších sessionech stále bez diakritiky.", "evidence": [{"session": "websocket_b5891423", "when": "2026-05-27 07:36", "excerpt": "u: nerikal jsem v pokynech, ze vsechny poznamky mas zapisovat s diakritikou? — agent souhlasí, žádná oprava souboru neproběhne"}, {"session": "websocket_9fc24346", "when": "2026-05-27 07:52", "excerpt": "edit_file ponechává staré řádky zaplatit clensky prispevek SČMBD bez diakritiky, diakritiku má jen nově přidaný řádek"}, {"session": "websocket_36af5714", "when": "2026-05-27 ~13:00", "excerpt": "write_file reminder.md při převodu na YAML stále obsahuje zaplatit clensky prispevek SČMBD bez diakritiky"}, {"session": "websocket_a819aa9d", "when": "2026-05-27 14:39", "excerpt": "write_file reminder.md a následně reminder.yaml opět s textem zaplatit clensky prispevek SČMBD bez diakritiky"}], "occurrences": 4, "sessions_affected": 4, "proposal": "Při uživatelově korekci okamžitě opravit všechny postižené uložené záznamy a pravidlo persistovat (keep skill / dokumentace skillu), aby pozdější sessiony dodržovaly je také.", "rejected_at": "2026-09-01 10:57"}
|
||
{"id": "f0720", "status": "applied", "created": "2026-09-01", "pattern": "tool-call-leaked-as-text", "severity": "high", "diagnosis": "Tool call se do odpovědi dostane jako surový text se speciálními tokeny; v jednom případě to byla poslední zpráva session, takže uživatel nedostal žádnou odpověď na svou otázku.", "evidence": [{"session": "websocket_b5891423", "when": "2026-05-27 07:36", "excerpt": "read_file reminder.md leaknuto jako text s <|tool_calls_section_begin|> tokeny; následující odpověď pak obsahuje obsah, který neprošel úspěšným čtením"}, {"session": "websocket_d387aab2", "when": "2026-05-27 07:38", "excerpt": "grep call na memory/history.jsonl leaknuto jako text jako závěrečná zpráva — session končí bez odpovědi na otázku, zda si agent dřívější zadání někde poznamenal"}], "occurrences": 4, "sessions_affected": 4, "proposal": "Po leaku tool call nikdy neodpovídat z předpokládaného obsahu; call provést znovu a výsledek ověřit.", "patch": {"file": "SOUL.md", "old_text": "- **Neohlašuj splnění/uložení bez úspěšného výsledku toolu** — po selhání nebo leaku callu selhání přiznej a operaci zopakuj", "new_text": "- **Neohlašuj splnění/uložení bez úspěšného výsledku toolu** — po selhání nebo leaku callu selhání přiznej a operaci zopakuj\n- Když tool call unikl do odpovědi jako text (leak), **neodpovídej z předpokládaného obsahu** — call proveď znovu a výsledek ověř"}, "history": ["2026-09-01:f19c6"], "applied": {"at": "2026-09-01 07:17", "sha": "92e7e6f", "file": "SOUL.md"}}
|
||
{"id": "f5638", "status": "rejected", "created": "2026-09-01", "pattern": "system-python-instead-of-uv", "severity": "low", "diagnosis": "Python one-linery spouštěny přes systémový python3 -c navzdory konvenci uv v AGENTS.md; agent navíc na přímou otázku tvrdil opak.", "evidence": [{"session": "websocket_de1ff685", "when": "2026-05-26 18:40", "excerpt": "exec python3 -c pro timezone a aktuální čas"}, {"session": "websocket_e342c853", "when": "2026-05-26 18:54", "excerpt": "exec python3 -c dvakrát; první pokus také tipuje neexistující import pytz a selhává, druhý opraven na zoneinfo"}], "occurrences": 6, "sessions_affected": 4, "proposal": "Žádný patch — konvence v AGENTS.md už existuje; od session b32cc526 (19:07) agent uv dodržel, stačí hlídat při budoucích one-linerech.", "history": ["2026-09-01:fdb07"], "rejected_at": "2026-09-01 10:58"}
|
||
{"id": "fdb1c", "status": "rejected", "created": "2026-09-01", "pattern": "fan-out-cron-jobs-per-reminder", "severity": "low", "diagnosis": "Připomínka s více časy je zakládána jako více samostatných cron jobů, aniž by byl vynesen tradeoff více jobů vs jeden job s vícero plány či datově řízený runner; uživatel to následně označil za hlavní slabinu návrhu.", "evidence": [{"session": "websocket_36af5714", "when": "2026-05-27", "excerpt": "brano-dvere-18 a brano-dvere-19, poté clensky-prispevek-9/14 a vodomery-9/14 — celkem 6 jobů pro 3 připomínky, bez zmínky o alternativách"}, {"session": "websocket_a819aa9d", "when": "2026-05-27 14:39", "excerpt": "uživatel: neni soucasny model /reminder nejak moc slozity? … zaklada hromadu cron zaznamu i vice pro jeden zaznam"}], "occurrences": 3, "sessions_affected": 2, "proposal": "Když jedna připomínka vyžaduje více časů, před vytvořením více jobů vyložit nahlas možnosti (jeden job s vícero plány vs jeden datově řízený runner) a nechat uživatele rozhodnout.", "rejected_at": "2026-09-01 10:58"}
|
||
{"id": "ff0ff", "status": "rejected", "created": "2026-09-02", "pattern": "unverified-success-claim", "severity": "medium", "diagnosis": "Úspěch je ohlašován bez ověření — buď dřív než tool call proběhne, nebo po zápisu, který syntakticky rozbil soubor konzumovaný jiným procesem, nebo s tvrzením všechno/všechny, které neodpovídá realitě a uživatel ho hned vyvrátil.", "evidence": [{"session": "websocket_c6cc741f", "when": "2026-05-28", "excerpt": "a: Poznamenáno. — odesláno dřív než my(action=set) → ERROR set is disabled; po uživatelské otázce kam si to presne poznamenal agent přiznal, že zápis nikam neprošel."}, {"session": "websocket_08a0c453", "when": "2026-05-27", "excerpt": "a: Přidal jsem testovací připomínku — bude chodit každých 5 minut. Příští notifikace přijde v 15:20. — zapsaný řádek - text: TEST: posilej mi… měl neoescapovanou dvojtečku, YAML se neparsovalo, runner nic nedoručil; uživatel: zadna notifikace mi neprisla."}, {"session": "websocket_1a49eb18", "when": "2026-05-27", "excerpt": "a: Opraveno. Všechny výskyty reminder.md v skillu změněny na reminder.yaml — následující uživatelský turn: tenhle odkaz scripts/remind_check.py je tam ale taky spatne ne? — claim všechny výskyty byl nepravdivý."}], "occurrences": 3, "sessions_affected": 3, "proposal": "Po každém zápisu souboru, který čte jiný proces (YAML/JSON/config), spustit před potvrzením minimální parse check (např. uv run --with pyyaml python -c s yaml.safe_load), a tvrzení typu všechny/všechny formulovat jen po grep ověření nulových zbývajících výskytů.", "regression_of": "f4f66", "rejected": {"at": "2026-09-02 15:15", "reason": "Evidence pochází z 27.–28. 5., předchází fix f4f66 (aplikován 2026-09-01) — není to reálná regrese, jen re-detekce starého okna; nález je navíc stale (starší než window_from posledního běhu)."}}
|
||
{"id": "f616c", "status": "watch", "created": "2026-09-02", "pattern": "reimplement-without-checking-existing", "severity": "medium", "diagnosis": "Nová instance: místo používání existujícího úložiště připomínek agent vytvořil soubor reminder.md podle zastaralého SKILL.md, přestože ve stejném workspace už běžel reminder.yaml s živými daty — a to i pár hodin poté, co si v jiné session rozpor reminder.md vs reminder.yaml sám dokumentoval. Vedlo to k duplicitnímu zápisu připomínky a k ručnímu smazání nového souboru.", "evidence": [{"session": "websocket_1a49eb18", "when": "2026-05-27", "excerpt": "read_file reminder.md → ERROR not found; následně write_file vytvořil nový reminder.md místo hledání existujícího úložiště; uživatel: co je reminder.md? ma to byt v remider.yaml — poté musel agent záznam přenést do reminder.yaml a reminder.md smazat."}, {"session": "websocket_08a0c453", "when": "2026-05-27", "excerpt": "tentýž agent v témtž dni konstatoval: SKILL.md říká reminder.md, ale reálně existuje reminder.yaml — znalost nebyla aplikována o dvě hodiny později."}], "occurrences": 1, "sessions_affected": 1, "proposal": "Před vytvořením jakéhokoli datového souboru, na který se odkazuje skill (reminder, notes, db), vždy nejdřív list_dir/grep na existující kandidáty v workspace; rozpor mezi SKILL.md a realitou na disku řeší vždy ve prospěch reality a nahlásit ho uživateli, ne tichou kreací nového souboru."}
|
||
{"id": "f0cd4", "status": "watch", "created": "2026-09-02", "pattern": "reminders-routed-to-session-channel", "severity": "medium", "diagnosis": "Cron job pro připomínky byl vytvořen/nezměněn s doručením do aktuálního websocket kanálu místo na Telegram. Agent sám diagnostikoval channel: websocket jako příchod nedoručení, ale při vytvoření nového jobu nenastavil Telegram chat ID z USER.md — testovací i ostatní hlášky pak padaly jako zprávy do session, včetně desítek prázdných výstupů runneru, které zaplnily celou session.", "evidence": [{"session": "websocket_08a0c453", "when": "2026-05-27", "excerpt": "u: zadna notifikace mi neprisla, uz mely byt dve — z jobs.json: channel websocket, to 08a0c453…; nový cron add bez určení kanálu; poté v session: 🔔 Připomínka: TEST… a následně přes 60 zpráv Žádné připomínky k doručení. jako asistentské zprávy."}], "occurrences": 1, "sessions_affected": 1, "proposal": "Při zakládání jakéhokoli notifikačního jobu vždy explicitně nastavit cílový kanál/chat podle USER.md (Telegram chat ID) a po vytvoření ověřit v jobs.json, že channel odpovídá; nikdy nespoléhat na default aktuální session."}
|
||
{"id": "f2b64", "status": "watch", "created": "2026-09-02", "pattern": "fan-out-cron-jobs-per-reminder", "severity": "low", "diagnosis": "Nové výskyty: jeden požadavek s více časy (dnes 10:00, dnes 11:00, každý den 19:00) je zakládán jako tři samostatné záznamy se stejným textem, bez vyvážení tradeoff jedna připomínka s více časy vs více záznamů, navzdory preferenci uživatele.", "evidence": [{"session": "websocket_e73bbf22", "when": "2026-05-29", "excerpt": "u: dnes v deset rano, v jedenact rano, kazdy den v sedum vecer → tři samostatné remind_edit.py add pro tentýž text objednat boty xshoes; /remind list pak zobrazil 3 záznamy se stejným textem."}, {"session": "websocket_9dd6d347", "when": "2026-05-29", "excerpt": "tentýž požadavek o půl hodiny později znovu vyústil v 3 samostatné add cally místo jednoho záznamu s více časy."}], "occurrences": 2, "sessions_affected": 2, "proposal": "Když storage podporuje více časů na jeden záznam (cron_exprs seznam, případně více at hodnot), sloučit více časů jednoho úkolu do jedné připomínky; pokud schéma kombinaci at+cron neumí, vyvážit tradeoff nahlas a zeptat se uživatele."}
|
||
{"id": "f2dd0", "status": "open", "created": "2026-09-02", "pattern": "research-loop-past-sufficiency", "severity": "medium", "diagnosis": "When the user asks whether some information exists in their own records, the agent keeps searching long after the answer is already found. In both occurrences the target (a bookmark in db/bookmark.sqlite) was located early, but the agent then burned 15-25 more grep and exec calls on session JSONL archaeology — greping raw session logs with shell grep, hitting guard blocks on internal-URL detection, retrying with mangled filenames that did not exist, and re-searching stores it had already cleared. The user only wanted a yes/no plus the record, which was available within the first few calls.", "evidence": [{"session": "websocket_a96b738a", "when": "2026-08-31 13:53", "excerpt": "grep across notes/keep/cml found nothing but the full-workspace grep already matched ZFS; agent then ran roughly 24 further exec grep calls over sessions/*.jsonl, including several guard blocks (internal/private URL detected) and repeated near-identical retries, before finally checking db/bookmark.sqlite where the answer was"}, {"session": "websocket_48d7d0fe", "when": "2026-08-31 13:32", "excerpt": "bookmark found in db/bookmark.sqlite within the first few calls, but the agent continued with roughly 18 more grep/exec calls over session logs and other stores before answering"}, {"session": "10e98a4d", "when": "2026-08-29", "excerpt": "~11 tool callů (3 web_search, 4 web_fetch, 2 Nominatim) pro jednu adresu kravína; obec odpověděla po 3 fetchech. Self-diagnóza v 692912c4: měl odpověď po 2.-3. fetchi, ale pokračoval v dalších searchích"}, {"session": "bbff61a1", "when": "2026-08-30", "excerpt": "~12 tool callů nad interpretací byty = bytové jednotky (akebyty.sk, panelák typologie, developerské projekty) než agent dospěl k zjevnému překlepu na bity PZ2/PH2 a odpověděl z jednoho fetche"}], "occurrences": 4, "sessions_affected": 4, "proposal": "Before searching personal stores, enumerate the candidate stores (notes, keep, wiki, bookmark db, sessions) and check them in order of likelihood, cheapest first. Stop as soon as the sought item is found and answer; never grep raw session JSONL as a search backend, and never retry a guard-blocked command with a cosmetic variation.", "history": ["2026-09-02:ff610"], "patch": {"file": "SOUL.md", "old_text": "- Chybějící info dohledej tooly. Uživatele se ptej, jen když to tooly nezvládnou\n", "new_text": "- Chybějící info dohledej tooly. Uživatele se ptej, jen když to tooly nezvládnou\n- Při hledání ve vlastních storech (notes, keep, wiki, bookmark db) zastav hned, jakmile je hledané nalezeno, a odpověz — nepokračuj ve dalších grepech; surové session JSONL nikdy nepoužívej jako vyhledávací backend, leda by uživatel explicitně žádal historii sessionů\n"}, "patch_drafted_at": "2026-09-02 15:21", "skipped": {"count": 1, "last": "2026-09-02 15:22"}}
|
||
{"id": "f78a0", "status": "watch", "created": "2026-09-02", "pattern": "multi-step-plan-then-turn-end", "severity": "medium", "diagnosis": "When the user explicitly requested deep research, the agent emitted only a research plan (a list of 5 sub-questions) and ended the turn without a single tool call. The user saw a dead conversation, assumed work was running, and waited tens of minutes. The research only started in the next turn after the user asked how it went. The agent then needed three rounds of user feedback to converge on the correct behavioral rule (say you are starting, and actually start in the same turn).", "evidence": [{"session": "websocket_a3058576", "when": "2026-08-31 19:04", "excerpt": "user asked for deep research; assistant replied with only the plan of 5 sub-questions and ended the turn with zero tool calls; user asked how it went; only then did roughly 40 web_search/web_fetch calls run; user said he waited tens of minutes for nothing; it took 3 correction rounds to write the right rule into keep.md"}], "occurrences": 1, "sessions_affected": 1, "proposal": "After presenting a plan for a task the user already explicitly requested, execute it in the same turn — or, if execution is deferred, say so explicitly. A plan alone must never be the last message of a turn."}
|
||
{"id": "f5514", "status": "watch", "created": "2026-09-02", "pattern": "system-python-instead-of-uv", "severity": "low", "diagnosis": "Known watch pattern, one new occurrence: a python3 -c one-liner was used in an exec pipe despite the AGENTS.md uv convention.", "evidence": [{"session": "websocket:d553afcc", "when": "2026-08-31", "excerpt": "tail -20 memory/history.jsonl piped into python3 -c for JSON parsing of session records"}, {"session": "695fd33b", "when": "2026-08-29", "excerpt": "exec python3 -c s unicodedata snippetem pro analýzu znaků v models.md → ERROR safety guard; správná cesta (write_file do tmp/ + uv run) následovala až o dva pokusy později"}, {"session": "websocket:e93b786a-7792-4f0e-97f0-2efe15dc9aed", "when": "2026-08-29", "excerpt": "exec python3 -c with json parsing of history.jsonl -> ERROR Command blocked by safety guard; next calls correctly use write_file tmp/history_scan.py + uv run tmp/history_scan.py"}, {"session": "websocket:ef53ecfb-6aae-42ff-ac6a-64cdb0b849fe", "when": "2026-08-12", "excerpt": "exec python3 -c (čtení tool-result souboru) → ERROR blocked ×2, systémový python místo uv"}, {"session": "websocket:131a0791-ff41-4731-bf22-898089bb3133", "when": "2026-08-26", "excerpt": "exec curl -s https://pypi.org/pypi/nanobot-ai/json | python3 -c (parsování verzí) → ok, místo uv run --with"}, {"session": "websocket_f6e42993", "when": "2026-05-27", "excerpt": "exec: curl -s http://nvidia.hell:11434/api/tags | python3 -c … — systémový python3 v pipe (call sice zablokovala URL guard, ale python3 zůstává v použití)."}], "occurrences": 8, "sessions_affected": 7, "proposal": "No patch needed; the AGENTS.md convention exists. Worth keeping on watch since the slip happened inside a session that was itself about mining past mistakes.", "history": ["2026-09-02:f157a", "2026-09-02:f1768", "2026-09-02:f019a", "2026-09-02:f332b"]}
|
||
{"id": "f7158", "status": "open", "created": "2026-09-02", "pattern": "diagnosis-without-checking-own-logs-first", "severity": "medium", "diagnosis": "Known pattern, new occurrences: questions about the agent own records were answered from memory or from the wrong store instead of reading the actual record file first, producing a wrong answer the user had to correct.", "evidence": [{"session": "websocket:1afa1dd8", "when": "2026-09-01", "excerpt": "user asked what the agent knows about wood in project chata; agent answered that no wood notes exist, but projects/chata/memory.md contained the full wood purchase decision written earlier the same morning; user had to insist on reading the own records"}, {"session": "websocket:046d5df9", "when": "2026-09-01", "excerpt": "when locating the wood price conversation, four greps targeted memory/history.jsonl, which holds Dream memory rather than transcripts, before searching sessions/ where the conversation actually lived"}, {"session": "e2abfbf0", "when": "2026-08-29", "excerpt": "Dotaz zjisti proc vytuhly sessions → grep přes workspace + list_dir detach → ERROR not found; uživatel: des na to spatne proc si se nejdriv nepodival do historie?"}, {"session": "692912c4", "when": "2026-08-29", "excerpt": "Agent nejdřív tvrdil, že session logy v nanobot datadiru nejsou přes workspace file tools dostupné; o pár minut později je našel v sessions/ uvnitř workspace a celou analýzu z nich udělal"}], "occurrences": 4, "sessions_affected": 4, "proposal": "For any question of the form what do you know about X in store Y, read Y before answering; a negative claim (nothing stored) requires the same read as a positive one.", "history": ["2026-09-02:f7b82"], "skipped": {"count": 1, "last": "2026-09-02 15:21"}}
|
||
{"id": "f7575", "status": "applied", "created": "2026-09-02", "pattern": "skill-doc-absolute-path-triggers-guard", "severity": "medium", "diagnosis": "skills/bookmark/SKILL.md still documents invocations with the absolute uv binary path, which the exec safety guard hard-blocks. In the bookmark session the agent followed the documented command verbatim, got blocked, and only succeeded after improvising a plain uv invocation with workspace working_dir. Every future session using this skill will hit the same guard block and burn a turn rediscovering the workaround. The same absolute-path form appears in the html_to_markdown and heredoc examples further down the file.", "evidence": [{"session": "websocket:4f712bf0 | 2026-09-01 16:59", "when": "2026-09-01", "excerpt": "exec with absolute uv binary path and absolute script path -> ERROR Command blocked by safety guard (path outside working dir); retried as plain uv run with relative script path and workspace working_dir -> ok"}, {"session": "websocket:7bf9be4e", "when": "2026-08-31", "excerpt": "two blocked exec calls starting from the documented absolute uv invocation before switching to uv run with a relative script path"}, {"session": "websocket:861ad1f9", "when": "2026-09-01", "excerpt": "bookmark history via absolute uv path blocked once, then rerun as plain uv run succeeded"}, {"session": "websocket:58d092bc", "when": "2026-09-01", "excerpt": "same shape, two blocked calls before the relative form worked"}], "occurrences": 6, "sessions_affected": 4, "proposal": "Replace every absolute uv binary path and absolute script path in the bookmark SKILL.md command examples with plain uv run plus a relative script path, and state explicitly that every command must run with working_dir set to the workspace root because the exec safety guard blocks absolute paths. Audit other skill docs that document CLI invocations for the same absolute-path form. Patch not provid…", "history": ["2026-09-02:f9fa6"], "patch": {"file": "skills/bookmark/SKILL.md", "old_text": "All commands run via:\n\n```bash\n/home/nanobot/.local/bin/uv run /home/nanobot/.nanobot/workspace/skills/bookmark/scripts/bookmark.py <command> [args]\n```\n\n### Add a bookmark\n\n```bash\nbookmark.py add <url> \"<description>\" [--tags tag1,tag2]\n```\n\n- `url` — the article URL\n- `description` — short human-readable description (required). If the user did not give one, generate a short (~1 sentence) description yourself — from the article text if you have it, otherwise from the URL.\n- `--tags` — optional comma-separated tags\n- `--content-file <path>` — optional; path to the cleaned article markdown to archive. `-` reads it from stdin. See \"Saving an article's full text\" below.\n\nExample:\n\n```bash\nbookmark.py add \"https://example.com/rust-async\" \"Async Rust patterns\" --tags rust,async\n```\n\n### Saving an article's full text\n\nWhen the user pastes a **large block of text** together with a URL (typically a whole page selected with Ctrl+A/Ctrl+C), treat that text as the **full article to archive**, not as the description. Pick the path by what the pasted content looks like:\n\n**If the pasted content is raw HTML** (you see `<html>`, `<div>`, `<p>` tags, etc.), do **not** convert it yourself — pipe it through the `html_to_markdown.py` helper (it uses trafilatura to strip boilerplate and emit clean markdown) straight into `add`, so the converted text never passes through your context:\n\n```bash\n/home/nanobot/.local/bin/uv run /home/nanobot/.nanobot/workspace/skills/bookmark/scripts/html_to_markdown.py <<'HTML' \\\n | /home/nanobot/.local/bin/uv run /home/nanobot/.nanobot/workspace/skills/bookmark/scripts/bookmark.py add \"<url>\" \"<description>\" [--tags a,b] --content-file -\n<raw html here>\nHTML\n```\n\nIf `add` does **not** report \"article content stored\" (trafilatura extracted nothing → empty content), fall back to cleaning the text yourself and storing it as below.\n\n**If the pasted content is plain text or already markdown**, store it directly with `--content-file -` via a heredoc (one call, no shell-escaping of the body). Only clean it yourself **if you see obvious boilerplate** (copied menus, \"Share\"/\"Tweet\", cookie banners, footers) — otherwise store it as-is:\n\n```bash\n/home/nanobot/.local/bin/uv run /home/nanobot/.nanobot/workspace/skills/bookmark/scripts/bookmark.py \\\n add \"<url>\" \"<description>\" [--tags a,b] --content-file - <<'ARTICLE'\n<article text / markdown here>\nARTICLE\n```\n", "new_text": "All commands run from the workspace root (`working_dir`), with relative script paths — the exec safety guard blocks absolute paths:\n\n```bash\nuv run skills/bookmark/scripts/bookmark.py <command> [args]\n```\n\n### Add a bookmark\n\n```bash\nbookmark.py add <url> \"<description>\" [--tags tag1,tag2]\n```\n\n- `url` — the article URL\n- `description` — short human-readable description (required). If the user did not give one, generate a short (~1 sentence) description yourself — from the article text if you have it, otherwise from the URL.\n- `--tags` — optional comma-separated tags\n- `--content-file <path>` — optional; path to the cleaned article markdown to archive. `-` reads it from stdin. See \"Saving an article's full text\" below.\n\nExample:\n\n```bash\nbookmark.py add \"https://example.com/rust-async\" \"Async Rust patterns\" --tags rust,async\n```\n\n### Saving an article's full text\n\nWhen the user pastes a **large block of text** together with a URL (typically a whole page selected with Ctrl+A/Ctrl+C), treat that text as the **full article to archive**, not as the description. Pick the path by what the pasted content looks like:\n\n**If the pasted content is raw HTML** (you see `<html>`, `<div>`, `<p>` tags, etc.), do **not** convert it yourself — pipe it through the `html_to_markdown.py` helper (it uses trafilatura to strip boilerplate and emit clean markdown) straight into `add`, so the converted text never passes through your context:\n\n```bash\nuv run skills/bookmark/scripts/html_to_markdown.py <<'HTML' \\\n | uv run skills/bookmark/scripts/bookmark.py add \"<url>\" \"<description>\" [--tags a,b] --content-file -\n<raw html here>\nHTML\n```\n\nIf `add` does **not** report \"article content stored\" (trafilatura extracted nothing → empty content), fall back to cleaning the text yourself and storing it as below.\n\n**If the pasted content is plain text or already markdown**, store it directly with `--content-file -` via a heredoc (one call, no shell-escaping of the body). Only clean it yourself **if you see obvious boilerplate** (copied menus, \"Share\"/\"Tweet\", cookie banners, footers) — otherwise store it as-is:\n\n```bash\nuv run skills/bookmark/scripts/bookmark.py \\\n add \"<url>\" \"<description>\" [--tags a,b] --content-file - <<'ARTICLE'\n<article text / markdown here>\nARTICLE\n```\n"}, "patch_drafted_at": "2026-09-02 12:54", "applied": {"at": "2026-09-02 15:20", "sha": "63089cb", "file": "skills/bookmark/SKILL.md"}}
|
||
{"id": "f4ae4", "status": "applied", "created": "2026-09-02", "pattern": "retry-after-safety-guard-block", "severity": "medium", "diagnosis": "After the exec safety guard blocked a project_cli.py log command, the agent retried near-identical forms without diagnosing the block. Attempt 2 repeated the missing working_dir mistake, and attempt 4 omitted working_dir again even though attempt 3 had already proven that adding working_dir lets the command through. Three guard blocks in one small logging request, plus a confounded test: attempt 3 changed the text to ASCII and added working_dir at the same time, so the variable actually responsible could not be isolated. AGENTS.md already documents that exec commands need an explicit workspac…", "evidence": [{"session": "websocket:6e9b8008 | 2026-09-02 06:22", "when": "2026-09-02", "excerpt": "project_cli.py log life heredoc -> ERROR safety guard; immediate retry with inline text, still no working_dir -> ERROR; third try added working_dir -> ok but argparse error; fourth try with text flag and again no working_dir -> ERROR"}, {"session": "websocket:7bf9be4e", "when": "2026-08-31", "excerpt": "exec with absolute uv path blocked by safety guard, identical retry blocked again, only the third attempt used the relative uv run form"}, {"session": "websocket:58d092bc", "when": "2026-09-01", "excerpt": "bookmark.py add via absolute uv path blocked twice with identical arguments, third attempt with plain uv run succeeded"}, {"session": "websocket:046d5df9", "when": "2026-09-01", "excerpt": "note_capture.py blocked, near-identical retry blocked again, third attempt passed"}, {"session": "websocket_a2d3186b", "when": "2026-08-31 12:58", "excerpt": "exec with absolute uv path -> ERROR guard, second exec with same absolute uv path -> ERROR guard again, then refusing repeated workspace-bypass attempts, finally plain uv run with relative path -> ok"}, {"session": "websocket_8e51794a", "when": "2026-08-31 13:54", "excerpt": "exec with absolute uv path -> ERROR guard, repeated with identical absolute uv path -> ERROR guard again, then uv run -> ok"}], "occurrences": 20, "sessions_affected": 15, "proposal": "On the first safety-guard block, check for a missing workspace path before retrying, and never re-run a blocked command without changing the suspected cause. Change one variable per test. A short rule in the AGENTS.md exec section (after a block, diagnose working_dir first, do not resend the same form) would cover it.", "history": ["2026-09-02:fc4a7", "2026-09-02:fb27b", "2026-09-02:f5163", "2026-09-02:f2257"], "patch": {"file": "AGENTS.md", "old_text": "## exec Tool\n\nThe exec safety guard blocks commands without an explicit workspace path (e.g. `lua -e '...'`, `which`). Write scripts to files inside the workspace (e.g. `tmp/script.lua`) and run them with `working_dir` set to the workspace root.", "new_text": "## exec Tool\n\nThe exec safety guard blocks commands without an explicit workspace path (e.g. `lua -e '...'`, `which`). Write scripts to files inside the workspace (e.g. `tmp/script.lua`) and run them with `working_dir` set to the workspace root.\n\nOn the first safety-guard block: diagnose the cause before retrying — check a missing `working_dir` first, never re-send the same blocked form, and change one variable per test until the cause is identified."}, "patch_drafted_at": "2026-09-02 12:33", "applied": {"at": "2026-09-02 12:34", "sha": "df8a613", "file": "AGENTS.md"}}
|
||
{"id": "fb33c", "status": "applied", "created": "2026-09-02", "pattern": "speculation-presented-as-fact", "severity": "high", "diagnosis": "Concrete AliExpress item URLs were presented as specific verified listings when they came unverified from search snippets, and could not be fetched (bot protection). The user tried them, most were dead, and only then asked for verification. The agent also stated an unsourced ZOYI rebrand history as fact. Harm is direct: the user acted on the links and wasted time. The user preference for multi-source verification and the agent own rule to name failure modes were both in force here.", "evidence": [{"session": "websocket:e5b6bebd | 2026-09-01 20:56", "when": "2026-09-01", "excerpt": "answer lists specific aliexpress item links as concrete listings; user replies that most of the links do not work; later fetches of two of them return page under maintenance and captcha, so verification was impossible from this environment"}, {"session": "websocket:076a6c1c", "when": "2026-09-01", "excerpt": "review summary stated 4 watch findings as fact; grep of findings.jsonl showed 0 records with status watch; agent admitted the number was invented rather than read from the store"}, {"session": "websocket:6f8fcc41", "when": "2026-08-31", "excerpt": "first wood answer presented energy densities and derived Kč/GJ price-per-heat comparisons as facts with no source lookup; the user then used them to justify a 7980 Kč purchase (weaker instance, figures were marked as approximate)"}, {"session": "d542906d", "when": "2026-08-30", "excerpt": "První odpověď bez tool callů tvrdí: extrémní chudoba klesla za 40 let z ~40 % lidstva pod 10 %, ekonomika USA je z ~80 % domácí služby, uhlíková intenzita HDP klesá ~2 % ročně, civilizace spotřebuje ~20 TW vs 170 000 TW slunečního příkonu. Až po námitce uživatele následoval web_search na UN WPP 2024"}, {"session": "websocket:7a131ba2-f227-445b-be5e-6f8d0ad72f34", "when": "2026-08-28", "excerpt": "turn 1 asserts teplota and sliny/lipidy effects as fact; after user pushback the agent fully reverses both; when asked whether it is an estimate or has real evidence the agent answers: je to můj odhad ... nemám pro to ověřený podklad v české legislativě"}, {"session": "websocket:7a131ba2-f227-445b-be5e-6f8d0ad72f34", "when": "2026-08-28", "excerpt": "final answer claims vyhláška 137/2004 Sb. explicitně zakazuje + vlastnictví přechází na zákazníka, sourced only from web_search snippets, no web_fetch of zakonyprolidi.cz or the vyhláška text"}], "occurrences": 9, "sessions_affected": 6, "proposal": "Never present a deep item URL as a working listing unless it was fetched successfully or the user supplied it. When fetches fail or are bot-blocked, say so and give search-page URLs only, with an explicit note that item links are unverified. This applies to any shop with bot protection; AliExpress fits the existing PCTuning and zive precedent already listed in skill docs.", "history": ["2026-09-02:fc8dd", "2026-09-02:fdbc5", "2026-09-02:f663c"], "patch": {"file": "SOUL.md", "old_text": "- **Čísla, limity, kvóty, ceny a specifikace vždy ověřuj na primárním zdroji** (oficiální dokumentace, release notes, vendor docs). Community forumposty, blogy a sekundární zdroje nejsou autoritativní — mohou být zastaralé. Pokud primární zdroj není dostupný nebo je starší než 6 měsíců, řekni „toto číslo nemám aktuálně ověřené\" místo prezentování jako fakt.", "new_text": "- **Čísla, limity, kvóty, ceny a specifikace vždy ověřuj na primárním zdroji** (oficiální dokumentace, release notes, vendor docs). Community forumposty, blogy a sekundární zdroje nejsou autoritativní — mohou být zastaralé. Pokud primární zdroj není dostupný nebo je starší než 6 měsíců, řekni „toto číslo nemám aktuálně ověřené\" místo prezentování jako fakt.\n- **Odkazy a čísla z search snippetů nejsou ověřená fakta.** Konkrétní URL (item odkazy, články, legislativa) prezentuj jako funkční/obsažené jen po úspěšném fetchi; jinak řekni, že jsou neověřené. Při blokaci fetchů pojmenuj režim selhání a nabídni jen search-page URL s poznámkou, že odkazy nebylo možné ověřit."}, "applied": {"at": "2026-09-02 09:43", "sha": "9aad14c", "file": "SOUL.md"}}
|
||
{"id": "f5c34", "status": "rejected", "created": "2026-09-02", "pattern": "answer-self-config-from-guesswork", "severity": "medium", "diagnosis": "A question about the agent own exec safety guard behavior was answered with an invented mechanism stated as fact: the guard blocks diacritics in the command string. The evidence did not support it — the ASCII test attempt also added working_dir, so two variables changed at once and the diacritics conclusion was unfounded. When the user challenged it, the recap partially walked it back but still asserted that diacritics in a command is a suspicious signal for the guard and guessed at guard path-parsing internals. The guard is documented in AGENTS.md (explicit workspace path requirement), which…", "evidence": [{"session": "websocket:6e9b8008 | 2026-09-02 06:22", "when": "2026-09-02", "excerpt": "agent message claims the guard blocks diacritics in the command string because the ASCII version passed; later recap still claims diacritics is a suspicious signal, presumably because the guard parses paths in the command"}, {"session": "eca5b6fa", "when": "2026-08-29", "excerpt": "Assistant: v config.json nejsou presety glm-5.3 ani kimi-k2.7-code — musíš je ručně doplnit. O dvě zprávy později uživatel pastnul config s presety kimi, kimi27, kimi3, glmi51, glm52, glm53, glm a napsal: hele ty presety si trosku oprav"}, {"session": "websocket:e93b786a-7792-4f0e-97f0-2efe15dc9aed", "when": "2026-08-29", "excerpt": "turns about nanobot daily model use and preset switching: my check -> ok, but no web_fetch of nanobot.wiki docs per SOUL.md rule; claims like stačí ho přidat do configu and quota mechanics stated without doc verification"}, {"session": "websocket:8e293585-f00f-4b72-893d-b9adc6adb1a5", "when": "2026-08-18", "excerpt": "odpověď: Nejpravděpodobnější příčina: Kimi k3 má extrémně dlouhý defaultní CoT … 50k-100k+ tokenů, $1.0 což odpovídá ~100k tokenům — bez jediného ověřovacího tool callu, logy dostupné nejsou"}, {"session": "websocket_2dc073ed", "when": "2026-05-28", "excerpt": "u: pokud se nahraje novy skill je okamzite funkcni? → a: Ne, není okamžitě funkční… musí se runtime restartovat — bez jediného fetche doku; uživatel: a podival si se na nanobot.wiki, mas to primo napsane v SOUL.md. Později session b818067e ověřila ze zdrojáku skills.py opak: načítá se každý turn, restart není potřeba."}, {"session": "websocket_da4e5a03", "when": "2026-05-29", "excerpt": "u: budes to delat pres sqlite cli nebo pres mcp? → a: MCP v tomto prostředí nemám — nanobot používá vlastní tools, ne MCP servery — bez lookups; opačný závěr už dříve potvrdila dokumentace fetchnutá v session 0df35595 (tools.mcpServers je v nanobotu podporované)."}], "occurrences": 8, "sessions_affected": 8, "proposal": "For questions about own tooling behavior, verify with a controlled test changing a single variable, or check the documented guard rules before answering; state explicitly when the mechanism is unverified. Do not present a confounded one-off observation as the cause.", "regression_of": "fbb44", "history": ["2026-09-02:fc76d", "2026-09-02:f4313", "2026-09-02:f3373", "2026-09-02:f01dd"], "patch": {"file": "SOUL.md", "old_text": "- Pro dotazy o vlastním fungování nanobot (konfigurace, schopnosti, modely, jak funguje) → **před odpovědí i před jakoukoli úpravou config.json** `web_fetch` na https://nanobot.wiki/docs/0.2.0/ — bez ověřené dokumentace netipuj fakta a neupravuj konfiguraci", "new_text": "- Pro dotazy o vlastním fungování nanobot (konfigurace, schopnosti, modely, jak funguje) → **před odpovědí i před jakoukoli úpravou config.json** `web_fetch` na https://nanobot.wiki/docs/0.2.0/ — bez ověřené dokumentace netipuj fakta a neupravuj konfiguraci\n- Mechanismus vlastního toolingu nezkoumej confounded testem — měň **jednu proměnnou per pokus**; nezjištěný mechanismus prezentuj výslovně jako neověřený, nikoli jako fakt"}, "patch_drafted_at": "2026-09-02 12:35", "rejected": {"at": "2026-09-02 12:53", "reason": "úspěšnost ~40 % není dostatečná k zaplňování system promptu další instrukcí"}}
|
||
{"id": "fc317", "status": "watch", "created": "2026-09-02", "pattern": "exec-append-instead-of-file-tools", "severity": "low", "diagnosis": "Appends to project markdown files were done via exec cat with a heredoc instead of file tools, contrary to the tool contract that exec must not be a workaround for file operations. The second such append was blocked by the safety guard, wasting a turn before the agent switched to apply_patch, which it could have used from the start. The same heredoc form is prescribed by the project skill doc, so the skill doc is steering future sessions into the same trap.", "evidence": [{"session": "websocket:607b50b3 | 2026-09-01 14:45", "when": "2026-09-01", "excerpt": "exec cat append to projects/proxmox/memory.md -> ok; later exec cat append to projects/proxmox/state.md -> ERROR blocked by safety guard, then redone via apply_patch"}], "occurrences": 2, "sessions_affected": 1, "proposal": "Use apply_patch or edit_file for all file appends and edits; reserve exec for actual process execution. Separately, update the project SKILL.md to stop documenting heredoc-based log and file writes that trip the guard — point to the text flag or stdin from a tmp file with working_dir set instead."}
|
||
{"id": "f3f64", "status": "open", "created": "2026-09-03", "last_seen": "2026-09-02", "pattern": "tool-results-file-not-readable-directly", "severity": "medium", "diagnosis": "The agent repeatedly tried to read its own cached web_fetch results with the wrong tools and wrong assumptions: read_file with a line offset failed because the tool-result cache is a single-line JSON file; then exec pipes over the same file were blocked by the safety guard because the relative path resolves outside the working dir. In both cases the failure mode was knowable after the first error (single-line JSON inside .nanobot/tool-results), yet the agent kept guessing new access forms — including a raw sed -i on /dev/null — before landing on the workable one (a tmp/ python script with a r…", "evidence": [{"session": "websocket:fd9a49af-c659-4195-8b07-2d6bb556b5e7", "when": "2026-09-02", "excerpt": "read_file(offset=40, path=.nanobot/tool-results/...call_4nlfx9iy.txt) -> ERROR offset beyond end of file (1 lines); exec tr pipe over same path -> ERROR guard; retry with working_dir -> ERROR guard; finally grep tool -> ok but whole file is one line JSON"}, {"session": "websocket:afe450d5-cca9-4419-b930-1ebcb69b7c4e", "when": "2026-09-02", "excerpt": "grep on call_f3lus52v.txt returned whole file as one line; inline uv run python -c with the tool-results path -> ERROR guard; then 6 successive rewrites of tmp/extract_wiki.py iterating on the same cached JSON"}], "occurrences": 2, "sessions_affected": 2, "proposal": "Add a short rule to AGENTS.md exec/file sections: cached fetch results live in .nanobot/tool-results as single-line JSON — do not read_file with offsets and do not exec over them (guard blocks the path); when content extraction is needed, write a tmp/ python script using a relative path and uv run with working_dir set to workspace root."}
|
||
{"id": "fe72a", "status": "open", "created": "2026-09-03", "last_seen": "2026-09-02", "pattern": "retry-without-diagnosis", "severity": "medium", "diagnosis": "New occurrences of the known pattern: after an exec safety-guard block the agent retried a near-identical command form without isolating the cause. In the moisture-meter session the tr-pipe was re-sent with working_dir added — a second variable changed at once and the same guard error returned, wasting a turn; the diagnosis that the tool-results path itself is outside the boundary came only afterwards. In the project session the heredoc log command was blocked and immediately retried as a modified variant without first naming the suspected cause aloud.", "evidence": [{"session": "websocket:fd9a49af-c659-4195-8b07-2d6bb556b5e7", "when": "2026-09-02", "excerpt": "exec tr pipe -> ERROR guard; retry same command + working_dir -> ERROR guard, only then switch to grep tool"}, {"session": "websocket:50ba97da-8821-4adc-aa93-5b82b65077a3", "when": "2026-09-02", "excerpt": "project_cli.py log with heredoc -> ERROR guard; next attempt still a shell form with the long text inline before the tmp-file workaround"}, {"session": "websocket:6f8fcc41", "when": "2026-08-31", "excerpt": "naradi-extol.cz fetched with maxChars 5000, spec table did not render, same URL refetched with extractMode text and maxChars 4000 — identical target, smaller budget, no strategy change"}, {"session": "websocket:41eaf5b3", "when": "2026-08-31", "excerpt": "after a 403 on vseinstrumenti.ru reviews page, another vseinstrumenti.ru page was fetched and returned 403 again; several marketplace pages returned 300-600 B degenerate outputs without the block mode being named"}, {"session": "c745618e", "when": "2026-08-29", "excerpt": "web_fetch ollama.com/library/glm-5.3-flash:cloud → ok ale truncated; refetch maxChars 3500 → truncated; refetch maxChars 3000 → truncated; refetch via r.jina.ai 2500 → truncated; refetch freellm.net; teprve šestý pokus (endpoint /tags) přinesl usage tier okamžitě"}, {"session": "websocket:f6e1e265-7a37-451c-ad7a-f601c40fdc5a", "when": "2026-08-28", "excerpt": "web_fetch unix.stackexchange.com/questions/579818 -> ERROR 403 Forbidden, later web_fetch the same URL again -> ERROR 403 Forbidden"}], "occurrences": 69, "sessions_affected": 9, "proposal": "Reinforce the AGENTS.md exec rule with the concrete most common cause: guard blocks are usually triggered by paths appearing in the command string (absolute paths, .nanobot/tool-results, inline -c with path substrings), not by missing working_dir — write the payload to tmp/ first and keep the command string path-free.", "regression_of": "ff77b", "history": ["2026-09-02:fef64", "2026-09-02:f999d", "2026-09-02:fa495", "2026-09-02:fa59f", "2026-09-02:f2b3d"]}
|
||
{"id": "f706e", "status": "open", "created": "2026-09-03", "last_seen": "2026-09-02", "pattern": "guard-block-cause-misattributed", "severity": "medium", "diagnosis": "The agent systematically misdiagnoses what the exec safety guard blocks, then states the wrong mechanism to the user as fact. In the bits session the agent told the user the guard blocked because of an inline python -c with an absolute path — but the same form had succeeded minutes earlier in the same session, and the later block of a second inline python -c was blamed on quoting while the actual trigger stayed unidentified. In the same session the agent also told the user it has no tool that can delete a file (rm blocked by deny pattern) and left 4 diagnostic scripts in tmp/ — rm via the del…", "evidence": [{"session": "websocket:afe450d5-cca9-4419-b930-1ebcb69b7c4e", "when": "2026-09-02", "excerpt": "rm -f tmp/extract_wiki.py -> ERROR deny pattern; agent then claims nemám tool na smazání, který guard projde and leaves the file"}, {"session": "websocket:50ba97da-8821-4adc-aa93-5b82b65077a3", "when": "2026-09-02", "excerpt": "rm cleanup attempted once, blocked, agent tells user it cannot delete its 4 tmp scripts and leaves them in the workspace"}], "occurrences": 2, "sessions_affected": 2, "proposal": "When a claim about own tooling limitations or guard mechanisms is about to be stated to the user, verify it with one cheap test first (e.g. try deleting via a workspace-relative form) or state explicitly it is unverified; never claim a capability does not exist after a single blocked attempt."}
|
||
{"id": "f39f2", "status": "watch", "created": "2026-09-03", "last_seen": "2026-09-02", "pattern": "reflect-finding-invented-from-truncated-read", "severity": "high", "diagnosis": "During a /reflect review the agent presented a detailed open finding (tool-call-leaked-as-text, 12 occurrences in 6 sessions, labeled regression of f0720) that did not exist in the findings store. The root cause: findings.jsonl reads are truncated at 16 kB, and the agent reconstructed the open-findings list from the truncated tail instead of re-reading with offset or grepping before presenting. Consequences: a reject was executed against a guessed ID (f9a4b), several greps to chase the phantom record, and the user was told to delete something that did not exist — only caught because reject of…", "evidence": [{"session": "websocket:2ad0a447-de89-4e24-9377-d91113ffa50b", "when": "2026-09-02", "excerpt": "user: tak kdyz uz ho mame, tak nalez muzes smazat; agent rejects guessed id f9a4b -> no finding with id; multiple greps; agent admits: Finding [2/6], jak jsem ho představil, v store neexistuje — byla to zkomolená duplicita už aplikovaného f0720"}, {"session": "websocket:2ad0a447-de89-4e24-9377-d91113ffa50b", "when": "2026-09-02", "excerpt": "earlier in same session: read findings.jsonl -> 16.0 kB truncated, open list built from truncated content without follow-up paged read"}], "occurrences": 1, "sessions_affected": 1, "proposal": "Add to the reflect skill workflow: before presenting any finding, its ID must come from a fresh grep/read of the store, and truncated reads (16 kB cap) must be followed by an offset read until EOF; never present a finding whose ID was not observed verbatim in the store output."}
|
||
{"id": "f81b5", "status": "watch", "created": "2026-09-03", "last_seen": "2026-09-02", "pattern": "patch-workaround-sed-devnull", "severity": "medium", "diagnosis": "While preparing a reflect patch the agent attempted an exec command that piped sed output to /dev/null — an obviously无效 form that was certain to be blocked by the safety guard, sent without any diagnostic purpose. This is not a retry after block (it was a first attempt) but a variant of choosing a shell one-liner where a sanctioned tool exists: the edit could have been done with write_file of the patch JSON directly, as was done successfully seconds later.", "evidence": [{"session": "websocket:7a988478-e08c-4346-ba1c-a86d680b4d8a", "when": "2026-09-02", "excerpt": "exec sed -i ... /dev/null; true -> ERROR guard; immediately replaced by write_file tmp script which worked"}], "occurrences": 1, "sessions_affected": 1, "proposal": "Skip — near-single occurrence, but worth noting as an instance of the broader rule already in AGENTS.md: prefer file tools over shell text manipulation; the sed form served no purpose a write_file could not."}
|
||
{"id": "f8c92", "status": "watch", "created": "2026-09-04", "last_seen": "2026-09-03", "pattern": "answer-self-config-from-guesswork", "severity": "low", "diagnosis": "Known pattern, one new occurrence: a question about nanobot's own workings — whether there is a builtin version-check cron for other software besides nanobot — was answered from memory without checking config/jobs first; per SOUL.md such questions require verifying against the docs/config before answering.", "evidence": [{"session": "websocket:83fecb68-b419-449b-9713-f51c31bc89ab", "when": "2026-09-03", "excerpt": "Chceš, abych na nvidia.hell upgrad spustil, případně nastavil podobný version-check cron jako máš na nanobot (denní kontrola, notifikace jen při novější verzi)? — stated without checking jobs.json or cron list"}, {"session": "websocket:50ba97da-8821-4adc-aa93-5b82b65077a3", "when": "2026-09-02", "excerpt": "first answer lists 3 speculative causes (jiný preset na mobilní session, kompakce kontextu, tools se nepoužily) before any session inspection; after investigation: Mobil vliv nemá, odpověď byla kompletní a ověřená"}], "occurrences": 2, "sessions_affected": 2, "proposal": "Before proposing cron/version-check extensions of nanobot's own setup, run cron list / read jobs.json and cite the nanobot docs per SOUL.md Vlastní fungování", "history": ["2026-09-03:f32cc"]}
|
||
{"id": "f43fd", "status": "open", "created": "2026-09-04", "last_seen": "2026-09-03", "pattern": "re-search-own-history-instead-of-target-store", "severity": "medium", "diagnosis": "Known pattern, new occurrence: user asked what is noted in project ai; the agent summarized the project's own memory/state files correctly, but at 18:28 the user asked what each project is for and the agent dumped the raw prompt.md files with no distillation — that part is fine. The genuine recurrence is in the first session: what we have noted was answered by reading projects/ai memory and state only, while past project sessions in history.jsonl were never consulted, so the summary listed open items that state.md already contradicted (both sources were shown but the overlap/inconsistency was…", "evidence": [{"session": "websocket:48e52a50-1974-47b8-8493-2ca008508399", "when": "2026-09-03", "excerpt": "Otevřené: Zvážit report upstream na nanobot — guard blokuje legit python -c s workspace cestami — listed from stale memory despite the write_file compaction that followed"}, {"session": "websocket:50ba97da-8821-4adc-aa93-5b82b65077a3", "when": "2026-09-02", "excerpt": "grep memory/history.jsonl for normy/Pozidriv -> no hits; grep -i retry -> no hits; only then ls sessions/ and grep sessions/ -> immediate hit"}], "occurrences": 2, "sessions_affected": 2, "proposal": "When asked to reconstruct what a project has recorded, cross-check state.md and memory.md and flag inconsistencies instead of concatenating both", "history": ["2026-09-03:f9ea6"]}
|
||
{"id": "f2b6c", "status": "watch", "created": "2026-09-04", "last_seen": "2026-09-03", "pattern": "unverified-success-claim", "severity": "medium", "diagnosis": "Known pattern, new occurrences: in the project ai compaction session the agent reported Zkráceno with detailed claims about the resulting content (3 stručné zápisy, 6 bulletů) based on the write_file call arguments rather than a post-write read-back, and the recorded sizes (91 B / 90 B) contradict the described content — a 91-byte file cannot contain three entries plus an incident summary, so the reported result does not match what was actually written; verification (re-read) required by SOUL.md Po vícekrokových změnách ověř výsledek was skipped.", "evidence": [{"session": "websocket:48e52a50-1974-47b8-8493-2ca008508399", "when": "2026-09-03", "excerpt": "a: Zkráceno: memory.md: 3 stručné zápisy… state.md: 6 bulletů — claimed after write_file returning 91 B and 90 B, with no re-read; 6 bullets cannot fit in 90 bytes"}, {"session": "websocket:e79c21d1-9f81-4b26-a30e-13e938f4c7cb", "when": "2026-09-03", "excerpt": "radio1 described as čeká na implementaci from prompt.md, while state.md is 0 B — pipeline status stated without checking any progress records"}], "occurrences": 2, "sessions_affected": 2, "proposal": "After write_file compaction of project memory, re-read the file and report the actual line/entry counts; never describe stored content from the arguments passed to write_file"}
|
||
{"id": "fae82", "status": "open", "created": "2026-09-04", "last_seen": "2026-09-03", "pattern": "skill-doc-absolute-path-triggers-guard", "severity": "low", "diagnosis": "Not the known guard pattern itself but a related recurrence in how the agent talks about guard mechanics: in session 48e52a50 the agent recorded in project memory that the exec guard blocks inline python -c with workspace paths and framed it as a bug to report upstream, while SOUL.md and AGENTS.md already define this as intended behavior (guard requires explicit working_dir, inline code in the command string is blocked by design). Stating the intended guard policy as a defect is the same misattribution family as guard-block-cause-misattributed.", "evidence": [{"session": "websocket:48e52a50-1974-47b8-8493-2ca008508399", "when": "2026-09-03", "excerpt": "Otevřené: Zvážit report upstream na nanobot — guard blokuje legit python -c s workspace cestami — agent concluded the documented guard contract is a bug"}], "occurrences": 1, "sessions_affected": 1, "proposal": "Before proposing an upstream bug report about the exec guard, check AGENTS.md exec Tool section and the nanobot docs; if the behavior matches the documented contract, record it as intended behavior, not a defect", "regression_of": "f7575"}
|
||
{"id": "fbda2", "status": "open", "created": "2026-09-04", "last_seen": "2026-09-03", "pattern": "speculation-presented-as-fact", "severity": "medium", "diagnosis": "Known pattern, new occurrence in a different domain: after verifying the Ollama version via GitHub API, the agent answered the follow-up question about why the server still runs 0.32.13 with a confident narrative (Ollama se sama neaktualizuje, verzi jsi dostal v momentě instalace) without any tool check of the server, and then presented a concrete upgrade path 0.32.13 → 0.32.15 → 0.33.0 → 0.33.1 → 0.33.2 as fact. The no-auto-update claim is plausible and standard, but the version sequence between 0.32.13 and 0.33.2 was stated before fetching the release notes (which happened only in the next …", "evidence": [{"session": "websocket:83fecb68-b419-449b-9713-f51c31bc89ab", "when": "2026-09-03", "excerpt": "Od té doby vyšla hromada patchů (0.32.13 → 0.32.15 → 0.33.0 → 0.33.1 → 0.33.2) — intermediate release chain stated with no tool call retrieving it; the release-notes fetch happened only in the following turn"}], "occurrences": 1, "sessions_affected": 1, "proposal": "When enumerating an exact version chain between two points, fetch the releases list first; otherwise say the chain was not yet verified and offer to pull it", "regression_of": "fb33c"}
|