battlecard
Skill0 starsChanged 4 months agoMITView on GitHub
Turn a tracked competitor's profile + live sources into a 6-section, rep-ready, fully-cited battle card. Triggers: 'battle card', 'battlecard', 'build a battle card for', '/battlecard'.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- calven-ai/calven-mcptheir own version, first seen Nov 1, 2026
- Hyde-and-Larkin/product-marketing-toolkittheir own version, first seen Nov 1, 2026
- guerrilla2799/sdr-copilottheir own version, first seen Oct 20, 2026
- calven-ai/marketing-as-codetheir own version, first seen Oct 7, 2026
- styfinity/claude-marketing-departmenttheir own version, first seen Jul 8, 2026
- GTMify/aigtmtheir own version, first seen May 25, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- GTMify/aigtm★ 25same name
- calven-ai/marketing-as-code★ 10same name
- styfinity/claude-marketing-department★ 6same name
- calven-ai/calven-mcp★ 3same name
- Hyde-and-Larkin/product-marketing-toolkit★ 2same name
- guerrilla2799/sdr-copilot★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 21
Not a research tool that invents claims.
- Distinctive: line 22
Not a one-time doc generator.
- Very specific: line 26
Read the competitor slug from the argument (e.g.
- Very specific: line 49
Write the card with exactly these sections.
- Very specific: line 60
Write the card to card_path using the exact structure in competitors/example-northwind/battlecard.md.
- Very specific: line 74
No API keys required for the skill itself - it uses Claude Code's built-in fetch + web search.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Battlecard
Generate a competitive battle card a rep can read in 90 seconds where every claim is cited to a source. The card is grounded, not guessed: each "where they win / where you win" line points to something the competitor actually published, so it survives contact with a skeptical prospect.
This skill reads a competitor profile you define once (competitors/<slug>/competitor.yaml), gathers evidence from that competitor's live sources, and writes a 6-section card to competitors/<slug>/battlecard.md.
Deploy when
- You're setting up a competitor for the first time and need the initial card.
weekly-refresh.pyflagged a card STALE because a source changed, and you're regenerating it.- A rep is about to walk into a competitive deal and wants the current card.
What this is NOT
- Not a research tool that invents claims. If a differentiator can't be tied to a source, it does not go in the card.
- Not a one-time doc generator. It's designed to be re-run every time a source moves. The card is disposable; the profile is the asset.
STEP 0: Parse input
Read the competitor slug from the argument (e.g. /battlecard northwind). Resolve competitors/<slug>/competitor.yaml. If it doesn't exist, stop and tell the user to cp -r competitors/_template competitors/<slug> and fill it in.
STEP 1: Load the profile
Parse competitor.yaml. Extract:
name,domain,categoryyour_product,your_differentiators(the case you want the card to make)their_known_strengths(seed honesty; you'll verify and expand these)sources[](label, url, type)card_path(where to write; defaultcompetitors/<slug>/battlecard.md)
STEP 2: Gather evidence
For each source in sources[], fetch the page and read it. Pay attention to source type:
changelog/docs-> what they shipped recently; capabilities they now claim.pricing-> tiers, what's gated, what's "contact us", per-seat vs. usage.reviews-> recurring complaints and recurring praise in the competitor's own customers' words.news-> funding, leadership, marquee customers, positioning shifts.
Also run a web search for "<name>" vs and <name> limitations to surface how the market already frames the matchup. Record the source URL next to every fact. You will cite it in the card.
STEP 3: Synthesize the 6 sections
Write the card with exactly these sections. Each factual line ends with a [label] citation matching a source.
- What they are - one neutral sentence.
- Where you win - exactly 3. Each must be specific ("no SSO below the $X tier [pricing]"), verifiable, and tied to
your_differentiators. No vague "we're more flexible". - Where they win - exactly 2. Honest, cited. This is the credibility of the whole card.
- Objections they create + your response - the top 3 objections a prospect will raise because this competitor exists, each with a calibrated 1-2 sentence response.
- Landmine questions - 3 questions the rep can ask that expose a weakness you found in Step 2.
- Competitive verdict - 1-2 sentences a rep can say out loud.
STEP 4: Write output
Write the card to card_path using the exact structure in competitors/example-northwind/battlecard.md. Start the file with a front-matter block recording competitor, generated (today's date), and the list of sources_cited.
End the card with a Sources list mapping each [label] to its URL, so any claim can be checked in one click.
STEP 5: Report
Print a 3-line summary: competitor name, how many claims are cited, and which sources were freshest (most recently changed). If any source in the profile was unreachable, say so explicitly. A card built on a dead source is a silent lie.
Output format
competitors/<slug>/battlecard.md, matching the example card's structure exactly: the 6 numbered sections, then a ## Sources block.
Configuration
No API keys required for the skill itself - it uses Claude Code's built-in fetch + web search. The refresh script (scripts/weekly-refresh.py) optionally reads SLACK_WEBHOOK_URL from the environment.
Installation
This skill ships inside the repo at .claude/skills/battlecard/. Opening the repo in Claude Code registers /battlecard automatically. To use it in another project, copy the battlecard/ folder into that project's .claude/skills/.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

