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/bks-lab/open-bridge/capability-brokernpx skills add bks-lab/open-bridge --skill capability-brokergit clone --depth 1 https://github.com/bks-lab/open-bridgeWrote 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/bks-lab/open-bridge/capability-broker)<a href="https://agentmods.dev/skills/bks-lab/open-bridge/capability-broker"><img src="https://agentmods.dev/badge/skills/bks-lab/open-bridge/capability-broker.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 | $0.00232 | $0.02005 |
| Opus 5 | $0.00116 | $0.01002 |
| Sonnet 5 | $0.00046 | $0.00401 |
| Haiku 4.5 | $0.00023 | $0.00200 |
Grade A, and why
capability-broker 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 6d 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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Capability Broker
The acquisition arm of the Bridge learning loop. The other arms look
backward and refine what exists (task-close-postmortem, bridge-audit,
bridge-curator); this one looks forward and gains a new capability — a
tool or a skill — when the user hits something the Bridge can't do yet.
It proposes, it never auto-acts. Every install, every outward step, every
persistent write is propose-then-confirm. Same human-gate-at-every-layer posture
as rules/learning-autonomy.md — this skill is bound by that rule, not an
exception to it.
Read this file only when triggered. Full procedures live in references/
(progressive disclosure) — load the one you need, not all four.
When this skill runs
| Entry | Mode |
|---|---|
The capability-gap-offer standing-order detects a true gap |
Auto-offer — surface the 4-way offer, then route |
| User says "the bridge can't do X", "build a skill for this", "is there a tool that…" | Direct — same flow, user-initiated |
User picks a route from a prior offer (r/s/o/d) |
Route — jump to the chosen branch |
The flow (one screen)
- Guard — confirm it's a true gap, not a false one. →
references/gap-detection.md - Offer — one question, four routes. Never silently fail or silently one-off.
- Route:
[r]research + provision a tool →references/research-playbook.md→ decision tree[s]scaffold a skill → decision tree → hand toskill-creator[o]one-off → do it now, ledger note only[d]decline → ledger note, cooldown- decision tree (provision vs skill vs both) →
references/provision-vs-skill.md
- Capture — write the learning so it compounds. →
references/learning-capture.md
1. Guard — true gap vs false gap
Before offering, run the cheap guard (full steps in
references/gap-detection.md). Five outcomes:
- T1 true gap — no skill, no wired tool, not native → offer (step 2).
- T2 dormant skill — a skill plausibly covers it but didn't fire → load it;
if it mis-triggered, log a
trigger-correction(existing loop). Stop — not this skill's job. - T3 native — Claude can already do it with built-in tools → just do it.
- T4 one-off — real but single-use → do it ad-hoc, ledger note.
- T5 declined/policy — declined this session or must-not-do → silent / refuse.
What ships with it
4 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.
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.
- 6d ago First seen · 163 lines · 232 tokens per session scan A 8997de219e27
capability-broker is a skill published in the GitHub repository bks-lab/open-bridge (8 stars, last pushed today), licensed MIT. It adds 232 tokens to every session and 2,005 once invoked, about $0.0012 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
xlsx
Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify…
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.
pptx
Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks.
docx
Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4)…
mcp-host-styling-integration
Integrates MCP App UI with host theming system. Applies host CSS variables, handles onhostcontextchanged, safe area insets, display mode detection, and fullscreen configuration.
quality-hooks
Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.