Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add danielwpz/pokoclaw --skill claude-session-observegit clone --depth 1 https://github.com/danielwpz/pokoclawWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/danielwpz/pokoclaw/claude-session-observe)<a href="https://agentmods.dev/skills/danielwpz/pokoclaw/claude-session-observe"><img src="https://agentmods.dev/badge/skills/danielwpz/pokoclaw/claude-session-observe/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/danielwpz/pokoclaw/claude-session-observe"><img src="https://agentmods.dev/badge/skills/danielwpz/pokoclaw/claude-session-observe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00037 | $0.01585 |
| Opus 5 | $0.00018 | $0.00792 |
| Sonnet 5 | $0.00007 | $0.00317 |
| Haiku 4.5 | $0.00004 | $0.00159 |
Grade A, and why
claude-session-observe scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Session Observe
Use this skill to inspect Claude Code's local session artifacts under ~/.claude.
The goal of this skill is not to teach you to trust a magic script. The goal is to teach you how to:
- find which Claude Code sessions exist for a project
- inspect what happened in a session
- understand what a session was doing from transcript evidence
- check whether work was completed, interrupted, delegated, or still needs follow-up
- compare the session story with the current repo state when the user wants to continue or verify work
Core principle
There are three layers here:
- Artifacts
~/.claudetranscript and sidecar files- current repo/worktree state for the recovered
cwd
- Query helpers
- optional scripts under
scripts/ - ordinary file reads,
ls,grep,tail-style inspection
- optional scripts under
- Your judgment
- "what sessions exist"
- "what this session was doing"
- "what was completed"
- "what should happen next"
The scripts are only layer 2. They are read-only retrieval helpers, similar to ls, tail, or select ... from ... limit .... You decide the final answer.
If the scripts do not exist, you should still be able to use this skill. The real contract of this skill is the investigation workflow.
Channels
- Active session map:
~/.claude/sessions/*.json - Project session transcripts:
~/.claude/projects/<encoded-project>/<session-id>.jsonl - Session sidecar directory:
~/.claude/projects/<encoded-project>/<session-id>/ - Subagent transcripts:
subagents/*.jsonl - Subagent metadata:
subagents/*.meta.json - Tool output captures:
tool-results/*.txt - Global history:
~/.claude/history.jsonl - Current repo/worktree for the recovered
cwd
Choose channels based on the question. Do not force every question through the same path.
Required first reads
- If the task is about
~/.claudelayout, path encoding, or where files live:- read
references/layout-overview.md
- read
- If the task is about finding which session to inspect:
- read
references/session-discovery-recipes.md
- read
- If the task is about how to read transcript JSONL records:
- read
references/transcript-records.md
- read
- If the task is about interpreting recent work or takeover / continuation:
- read
references/activity-heuristics.md - then read
references/continuation-handoff.md
- read
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/activity-heuristics.md 4.0 KB
- references/continuation-handoff.md 3.0 KB
- references/layout-overview.md 3.3 KB
- references/session-discovery-recipes.md 3.0 KB
- references/transcript-records.md 3.4 KB
- scripts/find_project_sessions.py 10 KB runs code
- scripts/read_session_records.py 15 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 179 lines · 37 tokens per session scan A 15b4c161ceed
claude-session-observe is a skill published in the GitHub repository danielwpz/pokoclaw (141 stars, last pushed 22d ago), licensed Apache-2.0. It adds 37 tokens to every session and 1,585 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
mem0-status
Diagnoses mem0 connectivity, API key validity, and memory read/write functionality. Use when memory operations fail, searches return empty, addmemory errors occur, or to verify the plugin is working correctly.
inspector-workbench
Runs Inspector UI work on the standalone Threads state lab so the agent can see the overlay. Use when a CopilotKit employee asks an agent to fix, polish, add, or debug Inspector UI, chrome, panes, overlay actions, launcher, Home, Threads, Playground, Memory, or any visual behavior in @copilotkit/web-inspector. Don't…
md-audit
Read-only code quality audit — scan the current working directory for common issues (bugs, dead code, security hotspots, missing error handling) and return a prioritised findings report. No files are edited. Use when asked to "audit the code", "quick audit", "find issues", "code scan", or "what's wrong with this…
potpie-debug-memory
Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history.
browserwing-admin
Manage and operate BrowserWing — an intelligent browser automation platform. Install dependencies, configure LLM, create/manage/execute automation scripts, use AI-driven exploration to generate scripts, browse the script marketplace, and troubleshoot issues.
exploiting-format-string-vulnerabilities
Methodology for exploiting format string bugs where attacker-controlled data reaches the format argument of printf-family functions, enabling stack/memory disclosure (info leaks for ASLR/PIE/canary defeat) and arbitrary write primitives (%n) to hijack control flow via GOT/.finiarray overwrites.