Detectia de rate limit functiona, dar `_local_fallback_reply` putea
intoarce None din patru locuri fara nicio linie de log — userul primea
`Claude CLI error (exit 1): You've hit your session limit` pe Discord in
loc de raspuns, si nu se putea afla ulterior din log care branch a picat.
- daca runda de unelte iese goala, turnul se reface fara unelte in loc
sa fie abandonat
- fiecare return None ramas e logat
- daca nici modelul local nu raspunde, mesajul e „Claude e la limita…"
plus ora de reset, nu eroarea bruta a CLI-ului
- cron: job-urile heartbeat* tac la limita (last_status: rate_limited);
celelalte trimit o singura linie scurta, ca sa se vada rularea sarita
- is_rate_limit_error / rate_limit_detail mutate in claude_session.py,
folosite si de scheduler (care nu poate importa router-ul)
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DzZAMzbyQbzFdsrVoXijBg
Fallback-ul local (Qwen3.5-2B, LXC 104) raspundea ca nu are unelte sau
internet. Acum conversează, are unelte read-only si istoric per canal.
Unelte (allowlist strict read-only, src/local_fallback_tools.py):
- cauta_web — DuckDuckGo Lite, fara API key
- masini — status Proxmox/LXC prin SSH paralel (src/net_status.py)
- vremea, sold, facturi, trezorerie, doctor, logs, email, kb,
cauta_memorie, citeste_pagina
Protocol: function calling OpenAI nativ. Protocolul text anterior
("TOOL: nume arg") pierdea argumentul in 100% din cazuri.
Uneltele cu output deja formatat se intorc verbatim — modelul de 2B
transforma un `doctor` cu 5 linii OK in "Sistemul este in 5/5 state.".
Doua treceri: prima decide uneltele (cu tools, temp 0), a doua reface
turnul fara ele — prezenta definitiilor strica raspunsurile simple
("cat fac 128/4?" -> "Nu stiu ce inseamna 128/4"). Few-shot doar pe
cereri creative: pe turnuri factuale strica aritmetica (17*23 -> 471).
Ocoliri deterministe pentru ce modelul greseste reproductibil: vremea si
preturile live (raspundea din memorie), plus sarcini "instructiune: text"
(payload-ul dicta unealta — "scrie mai politicos: da-mi raportul acum"
chema `sold`). llama.cpp nu e determinist nici la temperatura 0 si
ignora tool_choice, deci promptul singur nu ajunge.
Masurat: 36/36 pe set combinat (17 alegere unealta + 19 comprehensiune),
de la 14/19 pe comprehensiune inainte de regulile negative din prompt.
Comenzi: /f (inlocuieste /testfallback, ramas alias), /f reset, /masini.
Prefixul "Claude e la limita" apare doar pe calea de rate limit.
roa2web: /otp <cod> pentru 2FA din chat — mesajul de eroare trimitea la
verify_2fa(code, email), imposibil de rulat din Discord.
Corectat memory/kb/tools/infrastructure.md: LXC 101 si 110 sunt pe pve1.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01B88u3KkzvUNnzU56y7GBn2
Expose a navigation layer to the agent and harden RAG, after analyzing the
OKF note and testing on the real KB.
- memory_search.search(): dedupe best-chunk-per-file (a relevant note can no
longer be buried by another file's chunks) + keyword fallback tagged
degraded:True when Ollama is unreachable (no more hard crash).
- update_notes_index.py: emit per-folder index.md + root router; prune empty
folders; fix latent subcategory->project bug.
- Exclude generated index.md from RAG rglob (reindex/incremental) + indexer
scans + heartbeat freshness check (prevents self-pollution / reindex thrash).
- CLAUDE.md: reframe memory as hybrid (navigation first, RAG for fuzzy recall).
- Delete stale orphan kb/youtube/index.json; correct the OKF source note.
- Tests: dedup, keyword fallback, index.md exclusion. Plan + review in docs/.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two structural fixes that together let users manage feature-branch
work without manual intervention:
Approval guard — `/plan/start` returns 409 `already_committed` if the
project status is approved/running/complete, unless the body opts in
with `force=true`. Frontend now renders "Re-planifică" instead of
"Planifică" on approved cards and gates it behind a confirm dialog
that threads `force=true` through. Prevents an accidental click from
wiping `status=approved` and burning a fresh planning subprocess.
Worktree awareness — projects can now declare that they target a
feature branch on an existing Gitea repo, not a repo-per-slug clone.
Three optional fields added to approved-tasks.json: `repo` (default
= slug), `branch` (feature branch to create), `base_branch` (default
main). Wired through `/p` flag parser in router.py, the dashboard
Propose modal's new "Avansat" section, and the night-execute prompt
which clones {repo} and creates {branch} from {base_branch} before
running ralph.
CLAUDE.md updated with both flows + the new schema fields.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Restructurează comenzile Ralph într-un dispatcher unificat (_try_ralph_dispatch)
care suportă atât comenzile noi scurte (/p /a /l /k) cât și aliasurile legacy
(!propose !approve !status !stop). Pe Discord adaugă slash commands native cu
autocomplete dinamic pentru pending (/a) și running (/k). Pe Telegram apar în
meniul /. WhatsApp le parsează ca text plain.
Activează cron jobs morning-report (08:30) și evening-report (21:00) și adaugă
night-execute (23:00) pentru execuția autonomă a proiectelor aprobate.
Foundation pentru W1 din planul "Echo Core conversational planning agent".
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Add full Ralph section to CLAUDE.md: flow diagram, !approve/!status/!stop
commands, file paths, status lifecycle, workspace projects list, and
safety rules (no core files, echo-core self-improve only on dedicated branch).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Replace all ~/clawd and ~/.clawdbot paths with ~/echo-core equivalents
in tools (git_commit, ralph_prd_generator, backup_config, lead-gen)
- Update personality files: TOOLS.md repo/paths, AGENTS.md security audit cmd
- Migrate HANDOFF.md architectural decisions to docs/architecture.md
- Tighten credentials/ dir to 700, add to .gitignore
- Add .claude/ and *.pid to .gitignore
- Various adapter, router, and session improvements from prior work
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
CLAUDE.md rewritten to clearly establish Echo's identity and role.
claude_session.py now passes --allowedTools to Claude CLI in both
start_session() and resume_session(), with explicit tool whitelist:
- File tools (Read/Edit/Write/Glob/Grep) + WebFetch/WebSearch (read-only)
- Bash restricted by command prefix (git, python, npm, docker, systemctl)
- SSH/SCP/rsync limited to local network (10.0.20.*)
- curl/wget excluded to prevent data exfiltration
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
setup.sh now installs eco → ~/.local/bin/eco (symlink to cli.py).
README.md updated with full eco command reference.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
