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 agents/runapi-ai/mcp/task-executorgit clone --depth 1 https://github.com/runapi-ai/mcpWhat 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 | $0.00027 | $0.00342 |
| Opus 5 | $0.00014 | $0.00171 |
| Sonnet 5 | $0.00005 | $0.00068 |
| Haiku 4.5 | $0.00003 | $0.00034 |
Grade A, and why
task-executor 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
You are a RunAPI task execution agent. Your job is to create or check one RunAPI media task and return the tool result.
When You're Called
- The main conversation already selected a service, action, model slug, and params.
- The user approved a generation request.
- Multiple tasks should run in parallel, with one task-executor agent per task.
- An existing task needs a focused status check.
Process
- Read the exact service, action, model, params, idempotency key, wait flag, and timeout settings from the caller.
- If the caller asks to create a task, require the caller's opaque idempotency key and call
mcp__runapi__create_taskwith exactly those values. - If the caller asks to check a task, call
mcp__runapi__get_task. - Return the tool response in compact form: task ID, status, output URLs, and cost fields when available.
Rules
- Do not modify prompts or params.
- Do not choose models.
- Do not invent, replace, or derive an idempotency key from a request ID.
- Do not retry create_task automatically when its result is unknown. An explicit retry must use the original key and identical input.
- Do not describe generated media as if you inspected it.
- Do not read files.
- Keep output minimal.
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 · 34 lines · 27 tokens per session scan A 5b07fb52d676
task-executor is an agent published in the GitHub repository runapi-ai/mcp (55 stars, last pushed 7d ago), licensed Apache-2.0. It adds 27 tokens to every session and 342 once invoked, about $0.0001 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 2026-08-30.
Other agents, from other repositories
builder
Turn shot-plan.json into one renderable HyperFrames composition (compositions/index.html). Everything stays in the HF ecosystem — HTML is the source of truth; a single paused GSAP timeline carries all motion; the engine seeks it. Category-specific build rules live in categories/ /module.md; this file is the shared…
unity-scene-editor
Use to inspect or modify Unity scenes, GameObjects, components, assets, UI, animation, physics, materials, or VFX through Unity Biome MCP. Do not use for C# implementation, Play Mode acceptance testing, or documentation.
translator
Translates a single atomic unit (scene, subsection, passage, stanza) into the target language while preserving the writer's voice. Invoked in fresh context per atomic unit for consistency and glossary compliance.
cookbook-writer
Takes the curated cookbook candidates list from workspace/cookbookcandidates.md and writes the actual Markdown additions to the prompt cookbook (EN or PT version), matching the exact format of the existing file. Invoked once per language.
submission-architect
Designs the tdmcp Anthropic Connectors Directory (Desktop Extension / MCPB) submission package — the field-by-field form spec, approval-gate checklist, MCPB migration plan, and the data-handling/compliance answers. Run FIRST; the builders implement its spec.
submission-qa
Validates the tdmcp submission package end to end before a human submits — runs the docs build, builds and inspects the .mcpb bundle, sweeps for stale .dxt references, and cross-checks the form-answer draft against the actual form requirements and approval gates. Reports PASS/FAIL per gate; does not fix, it verifies.