Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/jackchuka/ghpmnpx agentmods add skills/jackchuka/ghpm/ghpm-suggestWrote 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/jackchuka/ghpm/ghpm-suggest)<a href="https://agentmods.dev/skills/jackchuka/ghpm/ghpm-suggest"><img src="https://agentmods.dev/badge/skills/jackchuka/ghpm/ghpm-suggest.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 5.1 | $0.00030 | $0.01329 |
| Opus 5 | $0.00015 | $0.00665 |
| Sonnet 5 | $0.00006 | $0.00266 |
| Haiku 4.5 | $0.00003 | $0.00133 |
Grade A, and why
ghpm-suggest 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ghpm-suggest
PREREQUISITE: Read
../ghpm-shared/SKILL.mdfor prerequisites and error handling.
Recommend what to work on next by assembling session context and project state, then reasoning about the best options.
Arguments
/ghpm-suggest— open-ended recommendation/ghpm-suggest <constraint>— any natural language constraint, e.g.:- "I have 2 hours"
- "I want to switch context"
- "I want to focus on frontend work"
- "something small"
Session Context
The following is auto-injected at invocation time. If any value is empty, that context is unavailable — skip it gracefully.
- Current repo: !
git remote get-url origin 2>/dev/null - Current branch: !
git branch --show-current 2>/dev/null - Recent commits: !
git log --oneline -5 2>/dev/null - Changed files: !
git diff --stat HEAD~5..HEAD 2>/dev/null - GitHub user: !
gh api user --jq '.login' 2>/dev/null
Workflow
Phase 1: Gather Context
-
Project config (required): Follow the startup sequence in
../ghpm-shared/SKILL.md. -
Project items (required): Load cache per
../ghpm-shared/references/cache.md. Focus on non-Done items: ReadyForDev, Planned, InProgress. -
Session context: Use the auto-injected values above. If GitHub user is available, find items assigned to them in cache. Find items in Done status assigned to user (these indicate completed work, but no completion date is available).
-
Relevant views (optional — skip if session context unavailable):
- If in a repo that matches a component, that component's view is relevant.
- Team standup views that match the user's assignments are relevant.
- Triage views are always relevant (unassigned work).
Degraded mode: If only project config and items are available (no git context, no user identity), suggest based purely on project state: prioritize ReadyForDev items with no assignee, then Planned items.
Phase 2: Reason and Suggest
- With all context assembled, reason about what to suggest. Consider:
- Proximity: items in the same repo/component as current work require less context-switching.
- Momentum: items related to recently completed work — the user has warm context.
- Status: ReadyForDev items are higher priority than Planned (already vetted).
- Dependencies: items unblocked by the user's recent completions (parent/sub-issue relationships if visible in cache).
- Size: if the user mentioned time constraints, prefer smaller items.
- Breadth: if the user wants to switch context, suggest items in different components.
- Assignee: prefer items already assigned to the user, then unassigned items.
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 First seen · 128 lines · 30 tokens per session scan A fe646b221b19
ghpm-suggest is a skill published in the GitHub repository jackchuka/ghpm (20 stars, last pushed 5mo ago), licensed MIT. It adds 30 tokens to every session and 1,329 once invoked, about $0.0002 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 skills, from other repositories
airtable-automation
Automate Airtable tasks via Rube MCP (Composio): records, bases, tables, fields, views. Always search tools first for current schemas.
accelo-automation
Automate Accelo tasks via Rube MCP (Composio). Always search tools first for current schemas.
acculynx-automation
Automate Acculynx tasks via Rube MCP (Composio). Always search tools first for current schemas.
aero-workflow-automation
Automate Aero Workflow tasks via Rube MCP (Composio). Always search tools first for current schemas.
affinity-automation
Automate Affinity tasks via Rube MCP (Composio). Always search tools first for current schemas.
apaleo-automation
Automate Apaleo tasks via Rube MCP (Composio). Always search tools first for current schemas.