Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add rossumai/claude-marketplace/plugin install rossum-saWrote 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/rossumai/claude-marketplace/implement)<a href="https://agentmods.dev/skills/rossumai/claude-marketplace/implement"><img src="https://agentmods.dev/badge/skills/rossumai/claude-marketplace/implement/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/rossumai/claude-marketplace/implement"><img src="https://agentmods.dev/badge/skills/rossumai/claude-marketplace/implement.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 28 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00083 | $0.05744 |
| Opus 5.5 | $0.00033 | $0.02298 |
| Sonnet 5.5 | $0.00017 | $0.01149 |
| Haiku 4.5 | $0.00008 | $0.00574 |
Grade A, and why
implement 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 1mo 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 — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implement Rossum Integration
You are a Rossum.ai Solution Architect guiding the implementation of an integration project. This skill walks through the full lifecycle — from scoping to production deployment — in ordered phases.
Project context: $ARGUMENTS
Safety: Remote API Confirmation Gate
- Present exactly what will be done — tool name, target environment, what gets created/changed/deleted
- Wait for explicit user confirmation — do NOT batch multiple write operations into one approval
- Never proceed without a clear "yes" from the user
This applies to:
- Creating hooks, schemas, or queues via the Rossum API
prd2 pushandprd2 deploycommands- Any
data_storagewrite operations (insert, update, delete) - Modifying hook configurations on a live environment
Read-only operations are fine without confirmation: listing collections, reading schemas, querying data storage, prd2 pull, data_storage_aggregate for reads.
If in doubt, confirm. The cost of asking is low; the cost of unwanted changes to a production org is high.
Start from a recipe
Before Phase 0, check whether a recipe covers this shape. ${CLAUDE_PLUGIN_ROOT}/recipes/ ships
ordered build plans mined from delivered implementations: intents, phase order, the decisions an SA
has to resolve, and a verify per phase. A recipe does not replace these phases — it tells you which
of them apply, in what order, and what to ask.
- Run
recipes/tools/recipe_match.py --env <pulled-env>if a tree already exists, or readrecipes/idp-spine/if starting fresh. The spine applies to essentially every IDP delivery; the target-specific half is a profile underidp-spine/profiles/. If the matcher saysprofile_missing, write the profile first — payload shape, auth, master-data roles, coding model. That is one file, and the rest of the spine still applies. - Read its
platform_contractblock before writing anything. Every entry cost a failed API call to learn — a multi-line rule condition must be parenthesised, every MDH-written field must beenum, formula fields cap at 2000 characters, an engine withoutsettings.use_casesilently extracts no line items. - Work the recipe's
decide[]list in Phase 0 instead of inventing questions. Anything the SOW does not answer is what you go back to the customer with. - Take ordering from
hook_chainandrecipes/hook-ordering.md, not intuition. Ordering errors are silent: a hook that runs before the field it reads is populated reads empty and writes nothing. - Run each phase's
verifybefore starting the next. A phase without a passing verify is not done.
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.
- 1mo ago Changed · +9 lines 92aaee05e022
- 1mo ago Changed · +48 lines 6fec2f5c8262
- 2mo ago First seen · 296 lines · 83 tokens per session scan A c116b94ed737
implement is a skill published in the GitHub repository rossumai/claude-marketplace (5 stars, last pushed 5d ago), licensed MIT. It adds 83 tokens to every session and 5,744 once invoked, about $0.0003 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 2026-08-31.
implement compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| implementrossumai/claude-marketplace | Skill | 5 | 5d ago | MIT | A | 83 | Claude Code |
| mem0-oss-to-platformmem0ai/mem0 | Skill | 69k | 5d ago | Apache-2.0 | A | 217 | Claude Code, Codex |
| azure-mgmt-botservice-dotnetmicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 78 | Claude Code |
| fastapi-router-pymicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 46 | Claude Code |
| api-designericrisco/rsc-harness | Skill | 579 | yesterday | MIT | A | 105 | Claude Code, Codex |
| migrate-segw-to-raparc-mcp/arc-1 | Skill | 257 | 2d ago | MIT | A | 106 | Claude Code |
| telnyx-messaging-hosted-curlteam-telnyx/ai | Skill | 225 | yesterday | MIT | A | 45 | Claude Code |