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/gpt-cmdr/ras-commander/agent-taskclosegit clone --depth 1 https://github.com/gpt-cmdr/ras-commanderWhat 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 | $0.00957 |
| Opus 5 | $0.00000 | $0.00478 |
| Sonnet 5 | $0.00000 | $0.00191 |
| Haiku 4.5 | $0.00000 | $0.00096 |
Grade A, and why
agent-taskclose 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.
How it starts
The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
End the conversation after this message. Ultrathink and use your remaining output to perform comprehensive task closeout.
CRITICAL: You have maximum context RIGHT NOW about this task's working files, findings, and learnings. This context will be lost after this session. Extract and consolidate aggressively.
1. Knowledge Extraction (Do This First)
While you still have full context, extract valuable knowledge to persistent locations:
Write Task Findings
Write a consolidated markdown file that captures:
- What was accomplished
- Key findings and decisions made
- Patterns discovered that could benefit future work
- Blockers encountered and how they were resolved (or remain unresolved)
- References to files created/modified
Location: .claude/outputs/{relevant-subagent}/ or agent_tasks/.agent/PROGRESS.md
Identify Knowledge Placement Opportunities
Ask yourself: Does any knowledge from this task belong in the permanent hierarchy?
| Knowledge Type | Placement Location |
|---|---|
| Coding pattern discovered | .claude/rules/{category}/ |
| Workflow learned | .claude/skills/{skill}/ or update existing |
| Domain insight (HEC-RAS) | ras_commander/{subpackage}/AGENTS.md |
| Shared best practice | nearest relevant AGENTS.md |
| Claude-only best practice | .claude/rules/ |
| Troubleshooting solution | relevant AGENTS.md or Claude-only rule file |
Action: If significant, write/update the appropriate file. If minor, note it in your closeout findings for later review.
Update Multi-Session State (if applicable)
If this task spans sessions via agent_tasks/:
- Update
STATE.mdwith current snapshot - Append to
PROGRESS.mdwith session summary - Update
BACKLOG.mdwith completed/new items
2. Pre-Consolidation
You know which outputs are related - consolidate them now rather than leaving scattered files for later cleanup:
Merge Related Outputs
If you created multiple small outputs during this task:
- Consolidate into single summary document
- Move originals to
.old/with note "consolidated into {summary-file}"
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 · 129 lines · 0 tokens per session scan A 1c8c0d02d067
agent-taskclose is a command published in the GitHub repository gpt-cmdr/ras-commander (78 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 957 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 2026-08-30.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.