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/rana/skills/recordgit clone --depth 1 https://github.com/rana/skillsWrote 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/rana/skills/record)<a href="https://agentmods.dev/commands/rana/skills/record"><img src="https://agentmods.dev/badge/commands/rana/skills/record.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 | $0.00000 | $0.00676 |
| Opus 5 | $0.00000 | $0.00338 |
| Sonnet 5 | $0.00000 | $0.00135 |
| Haiku 4.5 | $0.00000 | $0.00068 |
Grade A, and why
record 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.
What it actually says
Save cognitive output to the project's session archive with thread distillation.
Run after any skill or command produces output worth keeping. General-purpose capture — the explicit version of what dream and codex do automatically.
$ARGUMENTS
Process
-
Locate the output. Scan the conversation for the most recent substantial output — the last skill invocation's result, or the last analytical/creative block if no skill tag is obvious. If the argument names a specific skill or describes what to capture, use that as a hint.
-
Infer frontmatter. From conversation context, determine:
skill: which skill or command produced this (compose, crucible, invoke, gaps, etc.)chain: if compose, the full chain specfocus: subject matter (infer from the skill's arguments or the output content)threads: leave empty for now — populated in step 5converged: if a converge pass ran, its final verdict
If any field is ambiguous, make your best inference and note it.
-
Compress if needed. If the raw output exceeds ~400 lines:
- Preserve the full structure (section headers, key findings, verdicts)
- Compress supporting detail (evidence lists, intermediate reasoning)
- Keep all direct quotes and sharpest formulations verbatim
- Mark compressed sections with
[compressed]
If under 400 lines, save verbatim.
-
Save. Write to
dream/sessions/YYYY-MM-DD-HHMMSS.mdin the project root (not the skill repo). Createdream/sessions/if it doesn't exist. Use the current timestamp. Frontmatter block at the top. -
Thread check. Read
dream/INDEX.mdif it exists. For each active thread, assess: does this session's output touch it? If yes:- Add the thread name to the session's
threadsfrontmatter - Update the thread file: add a new evidence entry with session date and 1-3 sentence summary of what this session contributes
- If the session sharpens the thread's core question, update the "Current Sharpest Form" section
If the output introduces a tension not captured by any existing thread, note it but do NOT create a new thread. Threads require convergence from 2+ independent sessions. Instead, tag the session with
threads: [candidate:tension-name]so a future record pass can detect convergence.If
dream/INDEX.mddoesn't exist, skip thread check and note that the project doesn't have a thread archive yet. - Add the thread name to the session's
-
Update INDEX.md. If any threads were updated, refresh the session count in the INDEX table. If a thread was resolved by this session's findings, move it to the Resolved section with a one-line resolution.
-
Report. State:
- File path saved
- Which threads were updated (if any)
- Any candidate threads noted (if any)
- One sentence: what this session's output contributes that wasn't in the archive before
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 · 63 lines · 0 tokens per session scan A 84a02d42fae9
record is a command published in the GitHub repository rana/skills (1 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 676 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
learn
Force claude-smart to extract learnings from this session now.
init
Install the formatters this repository needs, with every command visible before it runs.
genshijin-compress
Markdown・テキストファイルを原始人形式へ安全圧縮.
standup
Show a daily standup summary with completed, in-progress, and blocked tasks across all active epics.
wp-polylang
Translate an existing WordPress site into a second language using Polylang.
check-dev
Type-check a Z specification with fuzz.