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 skills add delimit-ai/delimit-mcp-server --skill panelgit clone --depth 1 https://github.com/delimit-ai/delimit-mcp-serverWrote 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/delimit-ai/delimit-mcp-server/panel)<a href="https://agentmods.dev/skills/delimit-ai/delimit-mcp-server/panel"><img src="https://agentmods.dev/badge/skills/delimit-ai/delimit-mcp-server/panel/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/delimit-ai/delimit-mcp-server/panel"><img src="https://agentmods.dev/badge/skills/delimit-ai/delimit-mcp-server/panel.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00020 | $0.00293 |
| Opus 5.5 | $0.00008 | $0.00117 |
| Sonnet 5.5 | $0.00004 | $0.00059 |
| Haiku 4.5 | $0.00002 | $0.00029 |
Grade A, and why
panel 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 9d 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
Use delimit_deliberation_status and delimit_models with action="list" to identify the available route before running a panel. Tell the user if the route uses their configured model CLIs/API keys or hosted keys. A status check does not authorize a model call.
Frame one concrete question. Send only background context the user has approved for sharing with the panel. If the requested context is unclear, ask which details may be sent. Do not pass context_files unless the user approves each file: the engine reads the files and sends redacted, size-capped contents to models. Explain that the engine may also append local ledger, model, and git-status metadata automatically; see PRIVACY.md. If this is unacceptable to the user, do not call the panel.
Call delimit_deliberate with question and the approved context. Use mode, max_rounds, or scope only when they serve the request. Check for an error or blocked result before presenting a verdict. Report each model's actual position from the returned responses, where available, then state agreements, disagreements, and the tool's verdict. If the tool gives only truncated final responses, say so; do not invent missing positions or imply unanimity when status is no_consensus.
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.
- 9d ago First seen · 11 lines · 20 tokens per session scan A 3676dfe8ffb7
panel is a skill published in the GitHub repository delimit-ai/delimit-mcp-server (24 stars, last pushed yesterday), licensed MIT. It adds 20 tokens to every session and 293 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-08.
Other skills, from other repositories
API Contract Validator
Validate API responses against OpenAPI/Swagger specifications, JSON Schema definitions, and consumer-driven contracts to prevent breaking changes.
api-spectral
API specification linting and security validation using Stoplight's Spectral with support for OpenAPI, AsyncAPI, and Arazzo specifications. Validates API definitions against security best practices, OWASP API Security Top 10, and custom organizational standards. Use when: (1) Validating OpenAPI/AsyncAPI specifications…
coderifts
Call preflightchangeset before merge, deploy, publish, or tool registration when a contract artifact changed (OpenAPI/Swagger, GraphQL, gRPC/protobuf, AsyncAPI, MCP manifest, or agent tool schemas). Send the complete base-to-head change set. Branch on executionaction only. If you already hold a chain receipt and only…
coderifts
Before merging or shipping a change to an API or tool contract, preflight the change set, then branch on executionaction.
fastapi
FastAPI best practices and conventions. Use when working with FastAPI APIs, Pydantic models, dependencies, streaming responses including Server-Sent Events (SSE), and serving frontend apps. Keeps FastAPI code clean and up to date with the latest features and patterns.
printing-press-amend
Amend a published CLI from one of two input sources: (1) dogfood mode mines the active Claude Code session transcript for friction (missing flags, hand- rolled API payloads, silent-null returns); (2) direct-input mode accepts user-supplied asks (rename a command, add commands or feeds, fix a named bug, optionally…