Index of every page on this site: llms.txt
All skills
tristanmanchester avatar

/generating-novel-ideas

@3323bc9
by Tristan Manchestertristanmanchester/agent-skills3 stars
2
Run it yourself

Develop differentiated ideas for products, research, campaigns, names, or process changes. Use when the user asks to brainstorm, explore alternatives, escape generic concepts, or test a promising direction. Compare mechanisms and practical constraints; do not claim originality without an appropriately scoped search. Do not activate for proofreading, factual lookup, or implementing an exact spec.

Use this Skill: https://skilld.dev/gh/tristanmanchester/agent-skills/generating-novel-ideas

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesVALIDATION.md

≈363 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Validation

Local helper tests

From this skill directory run:

python -m unittest discover -s tests -v

These cover parsing, input bounds, token handling, and lexical candidate reporting. They do not evaluate the novelty, feasibility, or usefulness of generated ideas.

Agent evaluations to run separately

Compare complete traces with and without the skill for substantial product, scientific, campaign, naming-only, and process-redesign requests. Include a weak existing concept, conflicting constraints, and a request for exactly one idea.

Check whether the output addresses the actual brief; explores distinct mechanisms when exploration is needed; grounds current facts; separates proposals from evidence; handles the strongest objection; and proposes an ethical test tied to an assumption. Do not grade adherence to a fixed number of headings, personas, or contrast sentences.

Negative routing cases include proofreading, translation, summarising supplied text, factual lookup, and implementing an exact specification. A naming-only task should not trigger unsolicited product redesign. Same-context passes must not be reported as independent blind agent runs. A lexical-overlap score must not be reported as scientific or commercial novelty evidence.

Save the prompts, versions, outputs, reviewer judgements, and unresolved disagreements. The existence of these scenarios is not a claim that an agent evaluation has run.

Source: SKILL.md on GitHub

Skill behaviors

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

Third-party checks

No alerts1mo5 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The skill provides a framework for generating and evaluating novel ideas, utilizing a local Python script for lexical diversity analysis. The analysis script implements robust input validation, including strict size limits and data type checks. Dynamic module loading is observed in the test suite, where it is used appropriately to load the skill's own components and standard libraries for validation purposes. No security vulnerabilities or malicious patterns were detected.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

  • Runlayer7mo

    6 files scanned · No issues

  • ZeroLeaks6mo

    Score: 93/100 · 2 sections analyzed

Provenance

Signed by skilld at 3323bc9. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Steadyupdated last month

Capability

metadata
{
  "version": "2.1.0",
  "reviewed": "2026-09-13"
}
Other metadata
compatibility
No required external tools. Current market or technical claims need appropriate research. Optional lexical-overlap helper requires Python 3.10+ and no external packages.

README badge

README badge for tristanmanchester/agent-skills/generating-novel-ideas