Public contribution evidence can give a technical conversation a better starting point than a guessed title or a copied profile. This workflow uses Google Search to discover project pages, then records only explicit facts from an official repository or commit page. The sample shows public GitHub contribution evidence and deliberately leaves identity, contact, fit, and availability questions for human review.
A short source trail for a technical conversation, with concrete project evidence and no invented candidate record.
Source trail
01
vstinner
python/cpython
02
Joelgranados
torvalds/linux
03
Werkov
torvalds/linux
Editorial source audit example
Research workflow: public contribution evidence2026-09-17
Start with a brief
Turn one input into a usable research task
Find public contribution evidence for a technical conversation and preserve the source, role, artifact, and date without inventing contact data.
Stored locally for this task. It is not placed in the setup URL.
Generated prompt
Call spawnr_tools to inspect availability and the exact schema. The 17 September 2026 catalog check showed a $0 user price and required approval; the live quote is the authority for the execution cost ceiling. Pilot or daily allowances are runtime state, not a durable promise. Then call spawnr_tool_quote with {"tool":"google-search","input":{"q":"site:github.com/python/cpython/commit Python contributor buffer overflow","country":"us","language":"en","limit":10}}. Show the quote and wait for approval. If the tool is unavailable, the quote is unavailable, or its cost ceiling is unacceptable, stop before execution. Only after approval call spawnr_tool_run with the returned quote_id, approved=true, and the private access_token if returned; never display or export that private token. Treat returned links and snippets as discovery; ask the researcher to open official project or repository pages manually and record only explicit public evidence: account label, repository, author or committer role, contribution title, path, date, and source URL. Leave unknown identity and contact fields blank. Do not infer employment, seniority, ranking, fit, email, availability, or future activity.
Choose setup guide
Connect MCP opens the selected setup guide and carries this prompt with it.
Source: open sourceCaptured / reviewed 2026-09-173 records shown
What counts as evidence
The sample records three public repository pages. The CPython page identifies vstinner as the author of commit 9398655, titled “gh-156939: Fix xmlcharrefreplace() buffer overflow (#157109),” with a change in Objects/unicodeobject.c. A Linux page identifies Joelgranados as the committer of afdf35c, “sysctl: Fix type truncation in sysctl_msec_to_jiffies,” with a change in kernel/time/jiffies.c. Another Linux page identifies Werkov as author and htejun as committer for 057dac2, “cgroup: Avoid iteration of dying tasks with zero refcount.”
These are explicit page observations, captured 2026-09-17 from official GitHub repository pages. They show a public account label, project, contribution role, artifact, path, date, and URL. That is enough to prepare a precise question about the work. It is not enough to state where a person works, whether they are available, or whether they want to hear from you.
First-party project or repository page
Explicit author or committer label
Concrete commit, pull request, issue, or release artifact
Observed date and stable source URL
A source-first people research method
Begin with a project or technical topic rather than a demographic guess. Google Search can find public repository pages within a ten-result US/English snapshot. Open the result and read the project page itself. Keep the account label exactly as displayed, distinguish author from committer, and record the contribution title and file or project path when the page shows one.
Use the row to form a technical question: what trade-off did this change address, what context does the issue mention, or what adjacent work should you read? Then perform the fit check separately using the public profile, project documentation, or conversation context. Relevance alone does not establish seniority, current employment, or interest.
Search for public work, not private contact data
Copy explicit roles and titles only
Keep identity and fit as open questions
Let a human review the source before outreach
What this workflow does not do
The supported surface is Google Search, not a people database. It does not provide verified email addresses, phone numbers, a candidate database, employment verification, seniority scoring, inferred ranking, availability, outreach automation, or activity monitoring. The GitHub account label on a source page is not a verified legal identity. The rows are manual source examples and are not records returned by an enrichment or people MCP tool.
Respect the source boundary when exporting the table. Keep public URLs and explicit contribution facts. Do not copy email addresses that may appear in commit metadata, guess a work address, or turn commit frequency into a score. If a decision needs consent, current role, or contact details, ask through an approved human process and record the answer separately.
No verified emails or contact enrichment
No employment, seniority, or availability claim
No ranking or candidate fit score
No monitoring of future project activity
The next useful action
Search one project and one technical question. Open the official artifact, save a row with the source URL, and write a question grounded in the contribution itself. Before contacting anyone, have a human recheck the page and decide whether the proposed conversation is appropriate. The quality of this workflow comes from a traceable public fact and a careful next step, not from a larger list of names.
Search one project
Read the artifact
Ask a grounded question
Questions before you run it
Does this find people’s verified emails?
The supported task finds public project pages and records explicit contribution evidence. It does not find or verify email addresses, phone numbers, or private contact details.
Can a commit prove someone’s current job or seniority?
An account label and contribution role show what the project page records. They do not prove employment, title, seniority, availability, or interest in a conversation.
Why distinguish author from committer?
The official page can show different roles. Keeping “author” and “committer” separate preserves what the source says and avoids attributing a contribution more broadly than the page supports.
Can I rank candidates from this table?
The rows are source evidence for human review, not a ranking or fit model. Use the contribution to form a technical question, then evaluate relevance and consent separately.
What does the current Google tool catalog charge?
The 17 September 2026 catalog check reported a $0 user price and required approval. The live quote is the authority for the execution cost ceiling; pilot or daily allowances are runtime state and are not a durable promise. Stop before execution if the tool or quote is unavailable or the cost ceiling is unacceptable.