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 desplega-ai/agent-swarm --skill appsgit clone --depth 1 https://github.com/desplega-ai/agent-swarmWrote 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/desplega-ai/agent-swarm/apps)<a href="https://agentmods.dev/skills/desplega-ai/agent-swarm/apps"><img src="https://agentmods.dev/badge/skills/desplega-ai/agent-swarm/apps/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/desplega-ai/agent-swarm/apps"><img src="https://agentmods.dev/badge/skills/desplega-ai/agent-swarm/apps.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 | $0.00043 | $0.15260 |
| Opus 5.5 | $0.00017 | $0.06104 |
| Sonnet 5.5 | $0.00009 | $0.03052 |
| Haiku 4.5 | $0.00004 | $0.01526 |
Grade A, and why
apps 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 today.
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 — 750 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Swarm Apps
Swarm Apps are persistent, agent-authored internal applications: a schema-backed row store, named queries, custom script/task actions, and a validated json-render interface served at /apps/<id>. Use an app when people need to view and change live records; use create_page for a static snapshot.
Capability gate: app tools are available when the server's
CAPABILITIESincludespages(for exampleCAPABILITIES=core,task-pool,pages). There is no separateappscapability. If these tools are absent from your MCP list, this is why.
Tools and iteration loop
| Tool | Use it for |
|---|---|
app-list |
Discover apps by id, name, description, and timestamps. Definitions are omitted. |
app-get |
Read one complete stored definition before changing it. |
app-upsert |
Create an app, or replace an existing app's entire definition when you intentionally have the full desired state. |
app-patch |
Make a focused change while preserving everything not mentioned in the patch. Prefer this for iteration. |
app-query |
Run one declared named query and return its rows. Use this when an agent or saved script needs to consume app data rather than render the app UI. |
app-history |
List definition snapshots and their compact model/column digests before choosing a restore point. |
app-diff |
Show a unified definition diff; defaults to newest snapshot → CURRENT. |
app-rollback |
Restore a snapshot through the normal schema-migration engine. |
For every edit: app-get -> modify the smallest coherent subtree -> app-patch -> if rejected, fix every returned issues[] entry (path + message) and retry. Validation happens before storage, so a rejected update leaves the saved app unchanged. Do not guess at the current definition or use app-upsert with a partial definition.
app-patch can also change name; omit description to preserve it or pass description: null to clear it.
Reading an app's callable surface
One app-get returns everything needed to drive an app you did not write: definition.queries are runnable with app-query (supply every $param the query references), and definition.actions are invocable with POST /api/apps/<id>/actions/<name> (a script action runs its script, a task action spawns a task, a sync action refreshes sources). Read a query's $param placeholders and an action's kind/args from the definition before calling — do not guess parameter names.
What ships with it
2 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.
- today Changed · +7 lines 8f1e6f50140b
- yesterday First seen · 743 lines · 43 tokens per session scan A ed386070acf3
apps is a skill published in the GitHub repository desplega-ai/agent-swarm (890 stars, last pushed today), licensed MIT. It adds 43 tokens to every session and 15,260 once invoked, about $0.0002 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 2026-10-15.
Other skills, from other repositories
manage-task-checkpoints
Inspect and recover snapshots owned by a COSH managed Task. Use when the user asks the outer COSH Agent to list Task snapshots, preview or diff an exact recovery point, switch a Task workspace, or recover a workspace-only failure and retry once. Do not use for standalone ws-ckpt workspaces.
bbs-relay-pickup
A coordination procedure for agents that share work on tasks through a BBS task system, where BBS is the service used to claim, run, and report task steps.
security-observability
A read-only tool for examining security events already saved by agent-sec-cli, a command-line security log. It can connect those events to an agent session, run, or trace.
add-backend
Guide for adding a backend (Rust or Python) to the agent-sec-core security middleware. Use when creating new backends, integrating Rust or Python code into the security middleware, or extending with new backend actions.
skill-ledger
A security record for agent skills, which are reusable instruction packages for coding agents. It shows their status, reviews risk exposure, and can certify them with quick or user-approved deeper scans.
regex-mastery
Use this skill when writing regular expressions, debugging pattern matching,optimizing regex performance, or implementing text validation. Triggers on regex, regular expressions, pattern matching, lookahead, lookbehind, named groups, capture groups, backreferences, and any task requiring text pattern matching.