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 skills/gentbajko/capstone/corenpx skills add GentBajko/capstone --skill coregit clone --depth 1 https://github.com/GentBajko/capstoneWhat 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 | $0.00065 | $0.00187 |
| Opus 5 | $0.00032 | $0.00093 |
| Sonnet 5 | $0.00013 | $0.00037 |
| Haiku 4.5 | $0.00006 | $0.00019 |
Grade A, and why
core 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.
What it actually says
Capstone: core (internal)
This folder is the capstone suite's shared brain: references/
(core rules, protocols, craft files) and scripts/ (help, config
initializer, lint). Every other capstone skill resolves it as
../core/ from its own base directory, so install the whole suite;
a lone capstone skill without core cannot run.
Invoked directly, print the usage: run scripts/help.sh via bash
from this skill's base directory and output its stdout verbatim;
nothing else.
What ships with it
28 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/arch-craft.md 6.6 KB
- references/code-craft.md 8.0 KB
- references/core-authoring.md 5.5 KB
- references/core.md 20 KB
- references/dispatcher.md 1.1 KB
- references/interview.md 16 KB
- references/protocols/architecture.md 6.7 KB
- references/protocols/build.md 6.6 KB
- references/protocols/doctor.md 4.0 KB
- references/protocols/feature.md 3.1 KB
- references/protocols/groom.md 6.8 KB
- references/protocols/implement.md 11 KB
- references/protocols/logic.md 6.6 KB
- references/protocols/map.md 20 KB
- references/protocols/mockup.md 5.1 KB
- references/protocols/plan.md 6.4 KB
- references/protocols/review.md 9.5 KB
- references/protocols/stack.md 5.0 KB
- references/protocols/standards.md 6.1 KB
- references/protocols/start.md 3.6 KB
- references/protocols/uiux.md 14 KB
- references/style.md 1.8 KB
- references/topics.md 6.9 KB
- references/uiux-craft.md 21 KB
- scripts/help-hook.sh 1.4 KB runs code
- scripts/help.sh 3.7 KB runs code
- scripts/init-config.sh 8.9 KB runs code
- scripts/lint-sync.sh 27 KB runs code
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 · 17 lines · 65 tokens per session scan A 1fed48250261
core is a skill published in the GitHub repository GentBajko/capstone (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 65 tokens to every session and 187 once invoked, about $0.0003 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 skills, from other repositories
write-readme
Generates or updates a README.md for a project or package. Reads the source code to understand the project and produces a clear, accurate README following the project's conventions. Invoked when the user asks to write, update, or generate a README.
write-changelog
Generates a CHANGELOG.md entry for a release by summarizing git commits since the last tag. Groups changes by type: Added, Changed, Fixed, Removed. Follows Keep a Changelog format (https://keepachangelog.com). Invoked when the user asks to update the changelog, write release notes, or prepare a release.
project-map
Generates a compact structural orientation document (.agentic/project-map.md) so any agent can understand codebase structure without filesystem exploration. Invoked when onboarding to a codebase, exploring unfamiliar projects, or auditing code layout and conventions.
format-markdown
Formats Markdown files consistently using project-native tooling first (repo scripts/config), then safe fallback formatters, and verifies the result with lint/check commands. Invoked when the user asks to format markdown, tidy docs, normalize README files, or fix markdown style drift.
write-adr
Creates an Architecture Decision Record (ADR) documenting a significant technical decision. Follows the standard ADR format with context, decision, and consequences. Invoked when the user says "write an ADR", "document this decision", or "create an architecture record".
sync-confluence
Syncs engineering documentation (ADR, RFC, architecture) from docs/engineering/ to Confluence Cloud using the mark tool. Operates in two modes: scaffold (creates the directory tree locally) and CI sync (pushes on every main push). Only files with mark metadata headers are synced. Enforces diagram hierarchy (Mermaid >…