capability

A compatibility check that reports which capabilities a particular lens installation advertises to a programmatic caller. It reads the manifest and returns supported tokens, accepted input keys, and resolution rules.

In plain words
What is it for?
Checking whether an installed lens supports a requested capability and obtaining the manifest version and interface details a caller may use.
Why use it?
It lets an orchestrator check support before requesting a review, without dispatching work, reading code changes, judging them, or writing files.

Skill for Claude CodeCodex

Part of the lens plugin — 4 skills, 6 agents shipped together

Install

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.

agentmods
npx agentmods add skills/apurvbazari/claude-plugins/capability
Any agent
npx skills add ApurvBazari/claude-plugins --skill capability
Clone the repo
git clone --depth 1 https://github.com/ApurvBazari/claude-plugins

Made for: Claude Code, Codex.

Or install lens, the plugin that ships this one along with the rest of its 4 skills, 6 agents.

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,213 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00079 $0.03213
Opus 5 $0.00039 $0.01606
Sonnet 5 $0.00016 $0.00643
Haiku 4.5 $0.00008 $0.00321

Measured 2d ago against content hash fb7ce3bfb9c0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

capability 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.

lens/skills/capability/SKILL.md · 213 lines

How it starts

The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Capability — What This lens Advertises to a Programmatic Caller

You are invoked by an orchestrator (not a human), ahead of any call to lens:engine. You answer one question: what does this installation of lens support? You read one file, compare one list, and return one object. You dispatch no agent, read no diff, judge nothing, and write nothing at all.

The shape of both returns is declared in ../engine/references/engine-api.md § lens:capability. The advertised set — the tokens, the input keys, the resolution rules — lives here, because it is the answer rather than the shape.

Your input (supplied in context by the caller)

  • require — an optional array of capability tokens the caller needs. require absent ⇒ a pure report: you return the advertised set and can never return a capability error, because nothing was asked for that could be unsatisfied. Present and empty is the same pure report, for the same reason. A manifest that cannot be read still fails the call — that is an install-integrity failure rather than a capability miss, and it carries its own code (Step 1).
  • require present must be an array of strings. A bare string, an object, or an array holding a non-string is a caller shape bug, not a capability miss: return E_INVALID_INPUT (declared in ../engine/references/engine-api.md § Errors) before Step 2 runs. Never iterate a bare string — its characters are not tokens, and matching them would answer a question the caller never asked.

Step 1: Read the version from the manifest

Read .version from "${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json" — the double-quoted, plugin-root form — and report the filter's output verbatim:

jq -e -r 'if (.version|type) == "string" and (.version|test("\\S")) then (.version|sub("^\\s+";"")|sub("\\s+$";"")) else null end' "${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json"

Keep the double quotes. Unquoted, an unset or space-bearing ${CLAUDE_PLUGIN_ROOT} word-splits and the read dies as a command-not-found rather than as the file failure it actually is.

Read the full file on GitHub · 213 lines

Changes

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.

  1. 2d ago First seen · 213 lines · 79 tokens per session scan A fb7ce3bfb9c0

Subscribe to this mod's changes

capability is a skill published in the GitHub repository ApurvBazari/claude-plugins (0 stars, last pushed 26d ago), licensed MIT. It adds 79 tokens to every session and 3,213 once invoked, about $0.0004 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.