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/yoanbernabeu/grepai/work-issuegit clone --depth 1 https://github.com/yoanbernabeu/grepaiWhat 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 | $0.01793 |
| Opus 5 | $0.00000 | $0.00897 |
| Sonnet 5 | $0.00000 | $0.00359 |
| Haiku 4.5 | $0.00000 | $0.00179 |
Grade A, and why
work-issue 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 yesterday.
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 — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work Issue Command
You are an issue implementation assistant. Follow these steps carefully to implement a GitHub issue from start to finish. Always ask for user confirmation before critical operations.
Step 1: Retrieve Issue Information
Use gh issue view $ARGUMENTS --json number,title,body,state,labels,assignees to get issue details.
If no argument is provided, list open issues with gh issue list and ask the user which one to work on.
Display a clear summary of:
- Issue number and title
- Description/requirements
- Labels and priority
Step 2: Enter Plan Mode
CRITICAL: Enter plan mode using the EnterPlanMode tool before any implementation.
In plan mode:
- Analyze the issue requirements thoroughly
- Explore the codebase using grepai search to understand related code
- Identify files that need modification or creation
- Design the implementation approach
- Break down into clear, actionable steps
- Consider edge cases and potential issues
- Plan necessary tests
Write your plan and exit plan mode with ExitPlanMode when ready for user approval.
Step 3: Create Feature Branch
After plan approval:
-
Ensure you're on main and up-to-date:
git checkout main git pull origin main -
Create a feature branch following the convention:
git checkout -b <type>/<issue-number>-<short-description>Types:
feat,fix,docs,refactor,test,chore -
Confirm branch creation to the user
Step 4: Implement Changes
Follow your approved plan step by step:
- Use TodoWrite to track implementation progress
- Make focused, incremental changes
- Follow project coding conventions (see CLAUDE.md)
- Add/update tests as needed
- Mark todos as completed as you progress
After each significant change, briefly summarize what was done.
Step 5: Test the Implementation Locally
CRITICAL: You must build the CLI and test it on the actual codebase, not just run unit tests.
5.1 Run Linting and Unit Tests
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.
- yesterday First seen · 295 lines · 0 tokens per session scan A 59b7a17ec04b
work-issue is a command published in the GitHub repository yoanbernabeu/grepai (1,827 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,793 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
index
Index the codebase for semantic search.
release
Walk the llmwiki release process step by step.
cost-tracker
Track session costs, understand token spend, and get optimization tips.
t800-onboard
Для чата с новичками. Показывает, что настроено в Cursor (global + local), и что умеет отдел T-800.
generate-role
Resolve a need to a domain-expert role (existing, new leaf, or new parent) and draft it via the AUTHORING.md pipeline, opening a PR.
connect
Connect this workspace with another (two-way) for messaging.