hdresearch/vers-agent-services/log
Claude Code and agent skill by hdresearch/vers-agent-services
- community
Checking that the SKILL.md file is still in the repository.
Append-only work log for agent sessions. Use when recording what happened, what was decided, and what's next — the log is for the next session's orchestrator.
Install
npx vskill@latest install hdresearch/vers-agent-services/logSKILL.md
The same SKILL.md was published earlier as hdresearch/vers-agent-services/log.
Work Log
Carmack .plan-style append-only log. Every agent writes what they're doing as they do it. The log is the narrative record of a session — when a new orchestrator session starts, the first thing it reads is the log to understand what happened while it was gone.
When to Use
- Session starts — log that you're online and what you plan to work on
- LT dispatched — log which lieutenant was assigned which task
- Results arrive — log what a lieutenant produced (branch name, test results, key decisions)
- Decisions made — log why you chose approach A over B
- Problems hit — log what went wrong and how you resolved it (or didn't)
- Session ends — log a summary of what was accomplished and what's pending
The Key Rule
Write as things happen, not in batches. If you find yourself batching 3+ entries at once, you waited too long. The log should read like a timeline, not a retrospective.
Good: entry after each event as it happens. Bad: one giant entry at the end summarizing everything.
Who Is This For?
The log is for the next session's orchestrator. Assume the reader:
- Has no memory of this session
- Needs to pick up exactly where you left off
- Wants to know what happened, in what order, and why
- Doesn't want to dig through board tasks and feed events to reconstruct the timeline
The board tracks tasks. The feed tracks events. The log tracks the narrative — the human-readable story of what happened.
Convention
VERS_INFRA_URL env var points to the infra VM (e.g., http://abc123.vm.vers.sh:3000). All endpoints require Authorization: Bearer $VERS_AUTH_TOKEN.
API Reference
Append a Log Entry
Returns 201. Required: text. Optional: agent.
Each entry gets a ULID id and ISO timestamp automatically.
Query Log Entries (JSON)
Returns { entries: [...], count: N }. Entries are in chronological order (oldest first).
Query Log Entries (Plain Text)
Returns plain text, one line per entry:
The /raw endpoint is ideal for piping into a model's context at session start.
Common Patterns
Orchestrator Session Startup
Dispatching Work
Recording Results
Recording Decisions
Recording Problems
Session End Summary
Pi Tools
If the agent-services extension is loaded:
- `log_append` — Append a log entry
- `log_query` — Query entries with time filters
Log Entry Schema
Storage
Entries are stored as newline-delimited JSON (data/log.jsonl). Append-only — entries are never modified or deleted.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/log/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/hdresearch/vers-agent-services/log)