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/roomi-fields/rtfm/rtfm.ocrgit clone --depth 1 https://github.com/roomi-fields/rtfmWhat 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 | $0.00015 | $0.00396 |
| Opus 5 | $0.00008 | $0.00198 |
| Sonnet 5 | $0.00003 | $0.00079 |
| Haiku 4.5 | $0.00002 | $0.00040 |
Grade A, and why
rtfm.ocr 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 2d 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
The user wants to enable OCR for scanned PDFs in this project.
Run rtfm sync --ocr using the Bash tool. This command:
- Writes
"ocr_fallback": trueto.rtfm/config.json, so every future sync (CLI or auto-sync hook) automatically OCRs scans. - Invalidates the tracked hash of every PDF currently flagged in
.rtfm/seen_scans.json, so the next incremental sync re-ingests them through themarkerbackend. - Forks a detached background daemon (
rtfm ocr-worker) that does the actual OCR work. The command returns immediately with the daemon's PID — the user does not wait for OCR to finish at the prompt. - Is one-shot persistence: the user only runs this once per project. New scanned PDFs added later are OCR'd automatically by the next sync.
The daemon survives terminal close and Claude Code hook timeouts because
it is launched with start_new_session=True (its own process session,
immune to parent SIGHUP). If it crashes, .rtfm/ocr_state.json stays on
disk so the next rtfm sync --ocr resumes from where it left off.
After running the command, briefly tell the user:
- The PID of the started daemon (read from the command's output)
- That OCR runs in the background and will not block them
- That they can check progress with
/rtfm.statusorrtfm status
Do not wait or poll. The daemon's output is silent (DEVNULL) — the
status file is the source of truth, surfaced by rtfm status.
If the command output says "An OCR daemon is already running", report the existing progress to the user instead of relaunching.
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.
- 2d ago First seen · 34 lines · 15 tokens per session scan A 81c62ae7f0b3
rtfm.ocr is a command published in the GitHub repository roomi-fields/rtfm (24 stars, last pushed 2d ago), licensed MIT. It adds 15 tokens to every session and 396 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
rp-build-cli
Build with rp-cli context builder → chat → implement.
rp-investigate-cli
Deep codebase investigation and architecture research with rp-cli commands.
rp-oracle-export-cli
Export context for oracle consultation using rp-cli.
rp-reminder-cli
Continue your current workflow using rp-cli instead of built-in alternatives.
call-graph
Trace callers, callees, or paths using the call graph.
peek
Quickly find likely code locations without returning full code.