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 instructions/nklisch/skilltap/agents-mdgit clone --depth 1 https://github.com/nklisch/skilltapWhat 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 | $0.01307 | $0.01307 |
| Opus 5 | $0.00654 | $0.00654 |
| Sonnet 5 | $0.00261 | $0.00261 |
| Haiku 4.5 | $0.00131 | $0.00131 |
Grade A, and why
skilltap AGENTS.md 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 — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
skilltap
Product
skilltap is a personal control plane for managing supported agent harness environments. It adopts, reconciles, and updates native marketplaces, plugins, complete skill directories, MCP configuration, and shared instructions.
It does not provide discovery, recommendations, security scanning, a TUI, or compatibility with the previous implementation.
Foundation
Read these before designing or implementing changes:
docs/VISION.md— purpose, audience, principles, and exclusions.docs/SPEC.md— authoritative behavior, commands, state, and reconciliation.docs/ARCH.md— Rust architecture and module boundaries.docs/UX.md— non-interactive CLI behavior and output.docs/HARNESS-CONTRACTS.md— Codex and Claude native contracts.
Foundation documents describe current or intended truth. Planned and time-bound work belongs in .work/.
Product Constraints
- Commands are deterministic and non-interactive.
- Bare scoped commands operate globally.
--projecttargets the current project.--project <path>targets another project.--all-scopestargets the managed computer.--targetselects one registered harness or all enabled harnesses.~/AGENTS.mdis the canonical global instruction file.- A skill is the complete directory containing top-level
SKILL.md. - Native harness lifecycle commands take precedence over file materialization.
- Cross-harness transfer must be faithful or reported as partial.
- Partial foreground operations require explicit acknowledgment.
- Daemon updates never acknowledge partial operations or overwrite drift.
- skilltap does not search or browse marketplace contents.
Architecture
The implementation is a stable Rust workspace:
skilltap-coreowns domain types, state, planning, reconciliation, compatibility, and updates.skilltap-harnessesowns the typed target registry, distinct native harness adapters, and user-service adapters.skilltap-cliowns argument parsing, rendering, and exit codes.skilltap-test-supportowns fixtures and isolated test environments.
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 · 125 lines · 1,307 tokens per session scan A 38a40515652b
skilltap AGENTS.md is an instructions file published in the GitHub repository nklisch/skilltap (6 stars, last pushed 17d ago), licensed MIT. It adds 1,307 tokens to every session, about $0.0065 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 2026-08-31.
Other instructions, from other repositories
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
bread CLAUDE.md
Instructions for tjakoen/bread, covering claude.md — start here, what this is, start here (reading order), this repo is the stack's control plane — run it and non-negotiables (see conventions for the full rules).
getdesign AGENTS.md
Instructions for MohtashamMurshid/getdesign, covering agents, agent skills, issue tracker, triage labels and domain docs.
agentconfig.org AGENTS.md
Instructions for agentconfig/agentconfig.org, covering agent instructions for agentconfig.org, project overview, target audience, site structure and tech stack.
rosetta command-versions-vs-presence.instructions.md
Instructions for tikoci/rosetta, a project described as: MCP Server with RouterOS docs + commands + products + changelogs, using SQLite-as-RAG, sourced from MikroTik.
rosetta mcp-tool-descriptions.instructions.md
Tool descriptions should teach agents how to navigate the surface: workflow arrows, empty-result hints, boundaries, and result-budget cues.