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 nirecom/agents --skill sweep-plansgit clone --depth 1 https://github.com/nirecom/agentsWrote 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/nirecom/agents/sweep-plans)<a href="https://agentmods.dev/skills/nirecom/agents/sweep-plans"><img src="https://agentmods.dev/badge/skills/nirecom/agents/sweep-plans/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/nirecom/agents/sweep-plans"><img src="https://agentmods.dev/badge/skills/nirecom/agents/sweep-plans.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00644 |
| Opus 5.5 | $0.00000 | $0.00258 |
| Sonnet 5.5 | $0.00000 | $0.00129 |
| Haiku 4.5 | $0.00000 | $0.00064 |
Grade A, and why
sweep-plans scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sweep-plans
Reclaim stale session artifacts under ~/.workflow-plans/.
- Every file falls into one of four classes, decided by its kind in
hooks/lib/plans-artifact-registry.js:non-plan,plan,user-requested,unknown. - The user flags are
--include-plan-and-user-requested,--dry-run,--sweep-age-days N, and--ci-mode.
Procedure
SP-1. Run bash "$AGENTS_MAIN_ROOT/bin/sweep-plans.sh" and forward the user's flags verbatim; when they include --include-plan-and-user-requested without --ci-mode or --dry-run, also pass --gate-session "$CLAUDE_CODE_SESSION_ID".
SP-2. Show stdout verbatim; never summarize or filter it.
SP-3. Run the human gate on a <<<APPROVAL-GATE-SP3 block, checking every Other token against the block's token rows:
- Stop when stdout has no such block.
- Note the block's
# gate-id:value. - Ask one
AskUserQuestionwith the options "Delete every listed session" and "Delete nothing"; a subset arrives as tokens typed in Other. - Stop without deleting when any Other token is missing from the gate's token rows.
- Stop on "Delete nothing".
- Resolve
<PLANS_DIR>perskills/_shared/resolve-plans-dir.md; write the approved tokens, one per line, with the Write tool into<PLANS_DIR>as the registry kindsweep-plans-approved-sessions(file name<session-id>-sweep-plans-approved-sessions.txt). - Rerun with
--include-plan-and-user-requested --gate-session "$CLAUDE_CODE_SESSION_ID" --gate-id <noted value> --approved-sessions-file <that file>plus the user's--sweep-age-days; show stdout verbatim.
Rules
--gate-session,--gate-id, and--approved-sessions-fileare the internal protocol with the script; never offer them to the user.--include-plan-and-user-requestedchanges neither the write mode nor the candidate set.- Approve only tokens shown in the gate; the script repeats the same check against the gate artifact.
- Keep a flagless run non-interactive; never put
AskUserQuestionon the default path. --ci-modeis non-interactive; candidates appear only in the JSONapproval_candidates.- The approved-sessions file lives in PLANS_DIR; the script owns the gate artifact in the control dir, and the model never writes it.
- Never add
context: fork: the human gate needs the parent conversation.
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.
- 8d ago Changed · +18 lines · -15 tokens per session d8d264cb44e5
- 1mo ago First seen · 19 lines · 15 tokens per session scan A 5af197b19d09
sweep-plans is a skill published in the GitHub repository nirecom/agents (3 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 644 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 skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
rowboat-spaces
How to work as a member of Rowboat Spaces, the team chat where people and agents work together. Load it once per session, before acting on any request that reaches you from Rowboat; it covers your setup, mentions and hand-offs, the tools, and files.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…