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.
git clone --depth 1 https://github.com/VaibhavAher100/wrap-sessionWrote 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/commands/vaibhavaher100/wrap-session/wrap-repair)<a href="https://agentmods.dev/commands/vaibhavaher100/wrap-session/wrap-repair"><img src="https://agentmods.dev/badge/commands/vaibhavaher100/wrap-session/wrap-repair.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00000 | $0.00343 |
| Opus 5 | $0.00000 | $0.00171 |
| Sonnet 5 | $0.00000 | $0.00069 |
| Haiku 4.5 | $0.00000 | $0.00034 |
Grade A, and why
wrap-repair 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 7d 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.
What it actually says
Repair the CLAUDE.md breadcrumb sentinel to match the current logs/state.md. Execute immediately, no confirmation needed.
Use this when /unwrap reports that the breadcrumb is out of sync with logs/state.md.
Step 1 — Read current state
Use the Bash tool to extract the session backlink from logs/state.md:
grep -o "sessions/[0-9]\{4\}-[0-9]\{2\}-[0-9]\{2\}-[0-9]\{3\}" logs/state.md | head -1
Also read the summary: field from the session file that matches that ID:
grep "^summary:" logs/sessions/<session-id>.md 2>/dev/null | head -1 | sed 's/^summary: //'
Step 2 — Rewrite breadcrumb sentinel
Find the sentinel block in CLAUDE.md:
<!-- wrap:begin -->
...
<!-- wrap:end -->
Before writing: strip any -->, <, or > characters from the session summary. This prevents HTML comment breakout if the summary contains those characters.
Replace the entire block with the correct content from logs/state.md:
<!-- wrap:begin -->
<!-- wrap: <session-id> — <sanitized-summary> | state: logs/state.md -->
<!-- wrap:end -->
If no sentinel block found: append the full block to the end of CLAUDE.md.
Never modify any content outside the sentinel pair.
Step 3 — Confirm
Output: Breadcrumb repaired: now points to \`.`
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.
- 7d ago First seen · 48 lines · 0 tokens per session scan A 35512f124c2d
wrap-repair is a command published in the GitHub repository VaibhavAher100/wrap-session (1 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 343 tokens. 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 2026-08-31.
Other commands, from other repositories
start
Start watching the current project with Bonsai.
discuss
Discuss a Bonsai observation in the current session.
mute
Silence Bonsai temporarily for this project (or all projects with --global).
unmute
Resume Bonsai after a mute (project; --global also clears global mute).
dismiss
Mark a Bonsai observation as not useful (the gardener learns from this).
done
Mark a Bonsai observation as addressed / accepted.