New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Source Triangulation

by mohitagw15856 in mohitagw15856/pm-claude-skills

Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that…

MITAuto-check passedWriting & Content

Install Source Triangulation

skills CLI
$ npx skills add mohitagw15856/pm-claude-skills --skill source-triangulation -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install mohitagw15856/pm-claude-skills source-triangulation --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/mohitagw15856/pm-claude-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/source-triangulation .claude/skills/source-triangulation && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
source-triangulation
GitHub stars
1.4k
Token cost
~1.4k tokens
SKILL.md length
739 words
Files
1
Skills in repo
1,322
Repo updated
First seen
Licence
MIT

At a glance

Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that…

  • Works in 5 steps: Trace to the birth certificate: follow… → Independence is methods, not mastheads:… → Age and drift check: the origin's date… → …
  • Asked is this claim actually true
  • SKILL.md covers What This Skill Produces, Required Inputs, Framework: The Triangulation… and Output Format, plus 7 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Source Triangulation is an agent skill from mohitagw15856/pm-claude-skills. Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that separates established from echoed. Use when asked is this claim actually true, verify this stat before the deck, everyone cites this number where's it from, or how solid is this source. Produces the provenance trace, the independence assessment, the confidence grade with its reasoning, and the repeat-it-as phrasing.

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Writing & Content, covering Citation management and Copywriting. The repository describes itself as: 1255 professional Agent Skills for Claude, ChatGPT, Gemini, Cursor & Codex — PRDs, postmortems, leases, medical bills, layoffs, go-bags, new countries. Plain markdown, MIT, in… The licence is MIT.

When your agent uses it

  • Asked is this claim actually true
  • Verify this stat before the deck
  • Everyone cites this number wheres it from
  • How solid is this source

Example prompts

  • “/source-triangulation”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Trace to the birth certificate: follow each citation to its citation until the original appears — the survey, the paper, the filing, the…
  2. Independence is methods, not mastheads: two outlets citing the same study = one source; a survey and a separate dataset agreeing = two…
  3. Age and drift check: the origin's date rides with the claim ("a 2019 survey" is a different fact in 2026), and drift gets caught — claims…
  4. Grade honestly, four levels: established (2+ independent origins agreeing) · single-sourced (one origin, echoed widely — most business…
  5. Phrase at earned confidence: established → state it with the cite · single-sourced → attribute inline ("a 2024 Gartner survey found…") ·…

What it can do on your machine

Read from SKILL.md and the folder at commit 1cbf1f0. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Source Triangulation loads about 1.4k tokens when it runs. Until then it costs about 132 tokens; SKILL.md has 739 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~132
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from mohitagw15856/pm-claude-skills at commit 1cbf1f0, republished under its MIT licence (© mohitagw15856). 739 words, ~1,429 tokens.

Download SKILL.mdSave it as .claude/skills/source-triangulation/SKILL.md (or your agent's skills folder).
name
source-triangulation
description
Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that separates established from echoed. Use when asked is this claim actually true, verify this stat before the deck, everyone cites this number where's it from, or how solid is this source. Produces the provenance trace, the independence assessment, the confidence grade with its reasoning, and the repeat-it-as phrasing.

Source Triangulation Skill

Most "well-sourced" claims are one source wearing costumes: a stat born in a vendor's report, echoed by a blog, cited by a journalist, and now "widely reported" — three citations, one origin, zero verification. Triangulation is the discipline of asking two questions before repeating anything: where was this born (the provenance trace back through the citation chain to the original) and who has independently confirmed it (independence meaning different methods or data, not different websites). The output is a confidence grade and — the practical part — the phrasing that honestly matches it.

What This Skill Produces

  • The provenance trace — the citation chain walked back to the origin: who created this claim, how, when
  • The independence assessment — which supporting sources are actually independent vs. echoes of the same origin
  • The confidence grade — established / single-sourced / contested / unverifiable — with reasoning
  • The repeat-it-as phrasing — how to state the claim at its earned confidence ("one vendor survey suggests…" vs. the bare assertion)

Required Inputs

Ask for these if not provided:

  • The claim, precisely — "80% of projects fail" traced differently than "PMI found 80% of IT projects miss deadlines"; precision in the claim is precision in the trace
  • Where the user met it — the citing source starts the chain
  • The stakes — a deck stat, a strategy's foundation, a public claim? Depth scales: load-bearing claims get full traces; color commentary gets the quick check
  • The user's search access — the skill directs the trace; live searching (where available) executes it — otherwise the output is the trace plan with the checks to run

Framework: The Triangulation Rules

  1. Trace to the birth certificate: follow each citation to its citation until the original appears — the survey, the paper, the filing, the dataset. The questions at the origin: who made it, with what method, what sample, when, and with what incentive (vendor research isn't wrong by default, but its selection pressure is part of the record).
  2. Independence is methods, not mastheads: two outlets citing the same study = one source; a survey and a separate dataset agreeing = two. The echo test: do the supporting sources share numbers verbatim? Identical figures across "independent" sources is the fingerprint of a single origin echoing.
  3. Age and drift check: the origin's date rides with the claim ("a 2019 survey" is a different fact in 2026), and drift gets caught — claims mutate through citation ("IT projects over $1M" becomes "projects"; "survey respondents" becomes "companies"). The original's precise wording vs. the circulating version is often the finding.
  4. Grade honestly, four levels: established (2+ independent origins agreeing) · single-sourced (one origin, echoed widely — most business statistics live here) · contested (independent origins disagree — report the range, not your favorite) · unverifiable (no findable origin — the claim is folklore, however cited). The grade is the deliverable; flattering the claim helps nobody downstream.
  5. Phrase at earned confidence: established → state it with the cite · single-sourced → attribute inline ("a 2024 Gartner survey found…") · contested → the range with both cites · unverifiable → don't repeat it, or repeat it explicitly as unsourced folklore if rhetorically necessary. The phrasing rule is what makes triangulation usable — it converts the grade into words.
Show full SKILL.md (223 more words)Show less

Output Format

Triangulation: "[the claim]"

The Provenance Trace

[The chain: where met → cites → … → the origin (who, method, sample, date, incentive) — or the dead end, documented]

Independence Assessment

[Supporting sources sorted: independent (methods differ) vs. echoes (same origin) · the verbatim-number fingerprints noted]

The Grade

[Established / Single-sourced / Contested / Unverifiable] — [the reasoning, including drift found between original and circulating versions]

Repeat It As

"[The phrasing at earned confidence]" — [and the version for the deck's footnote]

Quality Checks

  • The chain was walked to an origin or an honest dead end
  • Independence was judged by method/data, not by outlet count
  • The origin's date and incentive are in the record
  • Drift between original and circulating wording was checked
  • The phrasing matches the grade — no confidence inflation in the repeat

Anti-Patterns

  • Do not count citations as confirmation — three echoes of one press release is one source with reverb
  • Do not stop at a reputable citer — reputable outlets echo single sources daily; the origin is the standard
  • Do not launder the grade in the phrasing — "studies show" for a single vendor survey is the exact sin
  • Do not pick a side in contested claims — the range is the honest fact
  • Do not repeat unverifiable claims bare — folklore travels on exactly that concession

Example Trigger Phrases

  • "Is this claim actually true?"
  • "Verify this stat before the deck."
  • "How solid is this source?"

© mohitagw15856, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/source-triangulation of mohitagw15856/pm-claude-skills.

Open the folder on GitHubat commit 1cbf1f0

Compare with similar skills

Source Triangulation next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Source Triangulation compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Source Triangulation this skillmohitagw15856/pm-claude-skills1.4k—~1.4kAutomated safety check: PassMIT
Biomedical Claim Checkerhh-health-AI/healthcare-equity101—~741Automated safety check: PassMIT
Fact Checkelvisun/newsjack1.5k—~5.6kAutomated safety check: PassMIT
Content Researchnicepkg/ai-workflow285—~3.7kAutomated safety check: PassMIT
Slackbot Builderswyxio/skills175—~2.1kAutomated safety check: PassMIT
Humanities Writing Companiontizzy916/humanities-writing-companion434—~3.4kAutomated safety check: PassCC-BY-NC-4.0

Similar skills

  • Biomedical Claim Checker

    hh-health-AI/healthcare-equity

    A skill your agent uses to audit whether medical or scientific claims are supported by their citations, verify references, assess AI-generated biomedical text or identify overstatement in a press…

    101 GitHub stars~741 tokensUpdated 3 days ago
    Writing & ContentAuto-check passed

  • Fact Check

    elvisun/newsjack

    Extract factual claims from PR copy, verify each claim independently, attach concrete citations, and warn when certainty is low.

    1.5k GitHub stars~5.6k tokensUpdated 2 days ago
    Research & ScienceAuto-check passed

  • Content Research

    nicepkg/ai-workflow

    Research-first content creation optimized for both human readers and AI search engines (Claude, ChatGPT, Perplexity, Gemini).

    285 GitHub stars~3.7k tokensUpdated 8 mo ago
    Writing & ContentAuto-check passed

  • Slackbot Builder

    swyxio/skills

    Design or harden Slack bot transport and interaction architecture: signed Events API ingress, fast acknowledgement, idempotency, causal thread context, Block Kit actions, message rendering…

    175 GitHub stars~2.1k tokensUpdated 4 days ago
    Writing & ContentAuto-check passed

  • Humanities Writing Companion

    tizzy916/humanities-writing-companion

    Help with humanities scholarly writing: sharpen research questions, map supplied readings, plan and draft arguments, review chapters, preserve authorial voice, and respond to reviewers.

    434 GitHub stars~3.4k tokensUpdated 17 days ago
    Writing & ContentAuto-check passed

  • Spark Report

    breakstageaxe61/genspark-claw

    Turn raw research, notes, links, or a conversation into a polished, shareable Sparkpage-style report page with an executive summary, structured sections, tables, and citations.

    141 GitHub stars~738 tokensUpdated 12 days ago
    Writing & ContentAuto-check passed

More from mohitagw15856/pm-claude-skills

All 1,322 skills in this repo

  • Exit Waterfall

    mohitagw15856/pm-claude-skills

    Compute who gets what at each exit price from a cap table — liquidation preferences, conversion points, and where the founders' share collapses.

    1.4k GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed

  • Feature Prioritisation

    mohitagw15856/pm-claude-skills

    Apply prioritisation frameworks (RICE, MoSCoW, Kano, ICE, Opportunity Scoring) to rank features and backlog items.

    1.4k GitHub stars~2k tokensUpdated yesterday
    Auto-check passed

  • Freelance Rate

    mohitagw15856/pm-claude-skills

    Derive a freelance day/hourly rate backwards from target income, honest billable utilization, overhead, and the self-employment tax premium — the arithmetic that proves a rate is not salary÷2000.

    1.4k GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed

  • Offer Comparison

    mohitagw15856/pm-claude-skills

    Compare two or more job offers as total-comp curves over four years — vesting cliffs, bonuses, 401(k) match, and the crossover year computed, not vibed.

    1.4k GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed

  • Refinance Breakeven

    mohitagw15856/pm-claude-skills

    Compute the month a refinance actually starts saving money — payment delta, breakeven month, and total interest on both paths including the term-reset trap.

    1.4k GitHub stars~1k tokensUpdated yesterday
    Auto-check passed

  • Rent Vs Buy

    mohitagw15856/pm-claude-skills

    Model rent-vs-buy honestly — year-by-year net position for both paths including the assumption everyone drops (the renter invests the difference), with a breakeven horizon instead of a verdict.

    1.4k GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Source Triangulation

What does Source Triangulation do?

Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that…. Source Triangulation is an agent skill from mohitagw15856/pm-claude-skills. Verify a claim before repeating it — the independent-sources test (three citations of one press release is one source), the provenance trace to the original, and the confidence grading that separates established from echoed.

When should I use Source Triangulation?

Source Triangulation fits situations like: asked is this claim actually true; verify this stat before the deck; everyone cites this number wheres it from; how solid is this source.

How do I install Source Triangulation in Claude Code?

Run `npx skills add mohitagw15856/pm-claude-skills --skill source-triangulation -a claude-code`. Or copy the skill folder (skills/source-triangulation in mohitagw15856/pm-claude-skills) into .claude/skills/source-triangulation in your project. Claude Code loads it when a task matches its description.

How do I install Source Triangulation in Codex?

Run `npx skills add mohitagw15856/pm-claude-skills --skill source-triangulation -a codex`. Or copy the skill folder (skills/source-triangulation in mohitagw15856/pm-claude-skills) into .agents/skills/source-triangulation in your project. Codex loads it when a task matches its description.

Can I use Source Triangulation in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mohitagw15856/pm-claude-skills --skill source-triangulation -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/source-triangulation, .gemini/skills/source-triangulation, .github/skills/source-triangulation and .opencode/skills/source-triangulation in your project.

What does Source Triangulation need to run?

SKILL.md names no scripts, command-line tools or credentials: Source Triangulation is instructions for the agent only.

Does Source Triangulation access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Source Triangulation safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Source Triangulation use?

Source Triangulation is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Source Triangulation use?

About 1.4k tokens (SKILL.md is roughly 5.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Source Triangulation?

Skills that share tags, products or a category with Source Triangulation: Biomedical Claim Checker (hh-health-AI/healthcare-equity, 101 stars), Fact Check (elvisun/newsjack, 1.5k stars), Content Research (nicepkg/ai-workflow, 285 stars) and Slackbot Builder (swyxio/skills, 175 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Source Triangulation?

mohitagw15856 (a GitHub user) maintains it in mohitagw15856/pm-claude-skills, which has 1,431 GitHub stars. The repository holds 1,322 skills in this directory. The repository was last updated on October 7, 2026.

Source: mohitagw15856/pm-claude-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.