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 agentmods add commands/conorbronsdon/agent-context-os/capturegit clone --depth 1 https://github.com/conorbronsdon/agent-context-osWhat 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 | $0.00016 | $0.00967 |
| Opus 5 | $0.00008 | $0.00483 |
| Sonnet 5 | $0.00003 | $0.00193 |
| Haiku 4.5 | $0.00002 | $0.00097 |
Grade A, and why
capture 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 3d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/capture — Triage Inbox
Read everything in inbox/, classify each item, propose where it goes, and route it only after you approve.
Instructions
1. Scan inbox
Read every file in inbox/ (skip README.md and .gitkeep). If nothing else is there, say so and stop.
2. Classify each item
For each item, determine its type and destination:
| Content type | Destination | Action |
|---|---|---|
| Task / TODO | TODO.md |
Append to the appropriate section |
| Decision or conclusion | state/decisions.md |
Add as a new entry (newest first) |
| Blocker | state/blockers.md |
Add with context |
| Priority change | state/current.md |
Update priorities |
| Writing idea or draft | relevant project context/draft file, or keep in inbox/ |
Move if ready, keep if raw |
| Reference link or note | relevant context file | Append to the appropriate section |
| ⚠️ Secret (password, PIN, seed phrase, recovery code, API key) | NEVER commit to the repo | Name the source file without reproducing its value; leave it in place and direct the user to move/remove it securely themselves |
| Unknown / multi-category | ask the user | Don't guess — present 1–2 options |
3. Present the triage plan (don't act yet)
Before moving anything, show the plan and wait for approval:
INBOX TRIAGE — [N] items found
1. [filename] → [destination] — [brief reason]
Destination write: [append/create and exact excerpt]
Source disposition: remains at inbox/[filename] until a separate user action
2. [filename] → [destination] — [brief reason]
Destination write: [append/create and exact excerpt]
Source disposition: remains at inbox/[filename] until a separate user action
[If any item is unclear:]
? [filename] — [description]. Suggested: [destination]. Correct?
Apply these destination writes? (or specify changes)
Do not move or edit files until the user approves.
4. Execute destination writes (after approval)
- Route each approved item to its named destination (append to existing files, or create new ones).
- Re-read each destination and verify the approved content is present before reporting success.
- Do not delete, move, truncate, or overwrite any source in
inbox/. - For a suspected secret, do not reproduce, copy, or relocate its value. Name only the source path and ask the user to move it to their password manager or delete it with a user-chosen secure action.
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.
- 3d ago First seen · 81 lines · 16 tokens per session scan A 88bb4c5de5a7
capture is a command published in the GitHub repository conorbronsdon/agent-context-os (17 stars, last pushed yesterday), licensed MIT. It adds 16 tokens to every session and 967 once invoked, about $0.0001 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 commands, from other repositories
recall-loop
Periodic memory digest — designed for use with Claude Code's /loop scheduler. Shows recent memories and pending to-dos in a compact, scannable format.
context-stats
Show context window usage, active tier, and system stats.
mode
Choose how this session may switch contexts on its own.
save
Save a Wingman handoff for the next tool or machine.
digest
What happened in a given week, verbatim.
timeline
Chronological history of this project's sessions, grouped by week.