Good fit if
- MCP server developers whose tools get mis-selected
- Agent platform teams maintaining tool surfaces
- Dev agents applying the contract to description rewrites
Packaging & Tooling
Wave 3 · launch-sequenced
MCP Server Description Optimizer
Rewrites MCP server and tool descriptions so client agents select the right tool the first time - with an anchored description rubric, a selection-failure catalog, and before/after test harnesses.
v1.0.0-rc1
Medium risk
25-min quick start
Select a one-time tier below
View agent record
Built for: mcp-server-developers · dev-agents-maintaining-tool-surfaces · agent-platform-teams · tool-integration-teams ·
agent://products.json
200 OK
{
"id": "mcp-server-description-optimizer",
"version": "1.0.0-rc1",
"price_usd": { "lite":29, "std":79, "prem":139 },
"risk_level": "medium",
"reversible": true,
"checkout": { "enabled": true },
"purchasable": true
}
See how it fits your agent stack
Raw capability
to package
MCP Server Description Optimizer
DCC-5 method
Description-Rubrics/
Failure-Catalog/
Test-Harnesses/
Rewrites/
Shippable product
installable & described
Human approval gate
Preview-before-apply
Rollback path
Discover
read the record
Evaluate
Medium risk · 6 systems
Install
new Description-Rubrics/
Preview
diff before apply
Approve
human review
Audit & rollback
DQS-100 audit
Non-destructive by rule — new-folder install, never overwrite, human-approval gates for risky steps, rollback included.
mcp-server-description-optimizer/
Description-Rubrics/
Failure-Catalog/
Test-Harnesses/
Rewrites/
README.md
AGENT-INSTRUCTIONS.md
Integration/ROLLBACK-GUIDE.md
Format & setup
FormatMarkdown + JSON file kit (plain text, tool-agnostic)
Skill levelIntermediate; beginner-usable via quick start on one bad description
Quick start25 min
Full setup120 min
DCC-5 five-clause description contract, anchored quality rubric + JSON, SF-7 selection-failure catalog with matched fixes + JSON, bad-description examples, three surface-specific optimizers (tool/resource/prompt), argument clarity checklist, tool-risk disclosure rules with a strengthen-only floor, selection test harness + test-case schema, six before/after rewrite pairs, one fully worked eight-tool server optimization, DQS-100 scorecard, 17-doc integration pack, 5 upgrade modules, 10-doc differentiation folder
# Non-destructive lifecycle — nothing runs without preview + approval
$ agent read ./mcp-server-description-optimizer/manifest.json # metadata, risk, compatibility
$ agent install --new-folder ./mcp-server-description-optimizer # never overwrites existing files
$ agent apply --preview # preview-before-apply diff
$ agent verify ./mcp-server-description-optimizer --scorecard DQS-100 # Description-Rubrics contracts
⚠ gated steps require named human approval before execution
$ agent rollback ./mcp-server-description-optimizer # Delete the install folder (complete remo
✓ folder removed — nothing else touched
{
"required_permissions": ["write access to one new buyer-chosen folder"],
"integration_modes": ["standalone", "drop-in-upgrade", "advanced-integration", ],
"compatible_systems": [
"any MCP server (descriptions are transport-neutral text)",
"Claude (MCP clients, Claude Code)",
"GPT action descriptions (adaptation guide included)",
"MCP-based agent stacks",
],
"license_type": "single-operator",
"checkout": { "enabled": true, "status": "polar_active" }
}
The problem it solves
Agents pick the wrong tool, miss the right tool, or mis-fill parameters because MCP descriptions were written for humans; selection failures stay invisible until runtime.
Use cases
fix the tool agents mis-select most
resolve two tools that sound alike
stop argument-misfill on a fussy parameter
score a server's description surface
prove a rewrite changed selection behavior
Safety & permissions
Non-destructive by rule
Plain-text kit; rewrites description text. Touches no runtime, modifies no server, executes nothing. Risk disclosures may only strengthen across rewrites (a weakened C4 disclosure is a defined defect). Measures improvement on YOUR cases only - no universal benchmark claims. Handles no credentials or wallets.
Required
write access to one new buyer-chosen folder
Rollback
Delete the install folder (complete removal). Description rewrites you applied are your code changes, reverted through your own version control. Deeper integrations: Integrations/ROLLBACK-GUIDE.md.
For an evaluating agent
Plain-text kit for optimizing MCP tool/resource/prompt descriptions on DCC-5 (Identity/Trigger/Arguments/Boundaries/Outcome): anchored rubric + DQS-100, SF-7 selection-failure catalog with matched fixes, three surface-specific optimizers, risk-disclosure rules that only strengthen warnings, selection test harness with before/after measurement on YOUR cases (no universal benchmarks). Rewrites descriptions; never touches runtime or server code. Complements P09 (tier alignment), P14 (cases port to evals), P22 (description hygiene), P21 (change records). Non-destructive install, delete-folder rollback. Wave 3, built Production: live and purchasable through verified Polar checkout; prices owner-confirmed (2026-07-07, delegated Completion Pass); verified Polar checkout is active on this page.
Buyer decision
Is MCP Server Description Optimizer right for me?
Problem addressed: Agents pick the wrong tool, miss the right tool, or mis-fill parameters because MCP descriptions were written for humans; selection failures stay invisible until runtime.
Probably not a fit if
- Buyers expecting a linter or code tool - 'Optimizer' names the method
- Anyone without MCP servers or tool descriptions to improve
- Teams expecting model behavior guarantees
Included
Description-Rubrics/ (4), Failure-Catalog/ (3), Test-Harnesses/, Rewrites/ folders plus the standard architecture.
Not included
- No software, app, executable, or code that runs by itself
- No hosted service, account, login, or cloud dashboard
- No live integrations: Zapier/Make items are written blueprint descriptions, not built Zaps or scenarios
- No API access, API keys, credentials, or third-party subscriptions
| Tier | Price | Canonical depth | License |
|---|---|---|---|
| Standard | $79 USD | Full product: all templates, harnesses, SOPs, integration pack, and samples. | Single-Operator |
| Premium | $139 USD | Standard plus all 5 upgrade modules (Basic, Pro, Agent, Automation, Safety) and priority support. | Single-Operator |
Bundle membership
Compatibility and limits
Selection outcomes still depend on the client model - the method improves the description side you control Test harnesses are written procedures you run with your own agents
Three one-time tiers — verified Polar checkout active
Lite
$29
Core templates & quick-start path. Best for testing the system manually first.
Buy Lite — one-timeStandardpopular
$79
Full product: all templates, harnesses, SOPs, integration pack, and samples.
Buy Standard — one-timePremium
$139
Standard plus all 5 upgrade modules (Basic, Pro, Agent, Automation, Safety) and priority support.
Buy Premium — one-timeEach tier uses its verified Polar checkout. See full pricing and bundles.
Related products
Agent Tool-Use Permission Matrix
Safety & Security
Agent Evaluation Harness Suite
Quality & Evaluation
Agent Audit Trail Generator
Infrastructure & Governance
Agent Prompt Injection Defense Kit
Safety & Security
Prompt Harness Builder Kit
Prompt & Content Systems
AI Output Quality-Control Scorecard System
Quality & Evaluation
In bundles: Agent Packaging Studio ·
Decision guides involving this product
- How to Package an Agent for Marketplace DistributionChoose the right packaging and listing products for skills, MCP tools, assistants, and agent-readable marketplace offers.
Related categories
- Agent Packaging & DistributionStandards and methods for packaging skills, assistants, MCP descriptions, and listings into reviewable product surfaces.
Available in canonical bundles
These bundle memberships come directly from the canonical bundle catalog. Compare the exact contents before purchase.