mcpbeat Sign in

Site Visit Report Skill for Claude

Create a typed, source-linked site visit report from field notes, photos, files, or conversation. Use for site observations, field reports, walkthrough records, existing-condition visits, and follow-up capture. Separate direct observations, participant-reported information, interpretation, limitations, issues, and proposed follow-ups. Saving a report never changes PROJECT.md, decisions/, or TASKS.md; promote only user-selected items afterward.

3k tokens
context cost
the whole folder, loaded on every use
3
files
instructions only
0
copies elsewhere
how many repositories repackaged it
302
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/AlpacaLabsLLC/skills-for-architects --skill site-visit-report

What comes with it

3 810 bytes besides the instruction
README.md
templates/site-visit-report.md

The instruction itself

10 sections, as written by the author

/as:site-visit-report — Typed Field Evidence

Create an auditable record of a site visit without turning limited observation, hearsay, or interpretation into verified professional conclusions.

Usage

/as:site-visit-report from field-notes/2026-07-21.md
/as:site-visit-report document today's walkthrough and attached photos
/as:site-visit-report revise site-reports/2026-07-21-roof-walkthrough.md

Hard rules

  • Classify every substantive item. Keep direct observations, participant-reported information, interpretation, issues, and proposed follow-ups separate.
  • State the observation boundary. Record access, visibility, weather, destructive-testing, equipment, and document limitations. An inaccessible or concealed area is Not observed, never “no issue observed.”
  • Do not claim concealed-condition certainty. Report only visible conditions within the stated visit scope. Do not infer what is behind finishes, underground, inaccessible, or otherwise concealed.
  • Do not claim compliance. Never declare code, zoning, life-safety, structural, MEP, accessibility, environmental, or contractual compliance. Route analysis to the appropriate specialist workflow and licensed professional.
  • Save before promotion. Complete and save the report before offering follow-up. Saving must not create or modify PROJECT.md, decisions/*.md, or TASKS.md.
  • Promote item by item. Require exact labels for every proposed fact, durable decision, or task; approval of the report as a whole is not promotion approval.
  • Preserve dates and provenance. Distinguish visit date, created date, source date, and photo/file metadata. Store project-relative links rather than absolute machine paths.
  • Never overwrite implicitly. Use deterministic numeric suffixes for collisions. Revise an existing report only when the user explicitly requests that exact revision.
  • Do not fabricate gaps. Unknown participants, locations, dates, conditions, photo identities, owners, or due dates remain Unknown, Not recorded, or Not observed.

Step 1 — Resolve the project root

Run the shared resolver and follow skills/project/references/context-resolution.md. Resolve exactly one validated project before reading sources or choosing a target. A studio-picker result requires one structured project-selection gate; do not ask first in prose. Stop on invalid, no-projects, or no-context. Resolve all inputs, the report target, and links relative to the selected project.

Step 2 — Establish visit scope, sources, and dates

Read the supplied notes, photos, drawings, files, and relevant conversation. Preserve each input as a project-relative source reference. For a conversational source, record Current conversation (not separately archived).

Determine independently:

  • Visit date: when the visit occurred. Prefer an explicit source or user date. Ask when unavailable because it controls the canonical filename.
  • Created date: when this artifact is written.
  • Visit purpose and scope: why the visit occurred and which locations or systems were intended for observation.
  • Participants: people present and their roles, only when supported.
  • Conditions: weather, lighting, occupancy, access, or active work that materially affected observation.

Never infer the visit date from file metadata. Treat photo timestamps and filesystem modification times as metadata, not proof of the visit date.

Step 3 — Classify the evidence

Assign stable labels and do not merge categories:

  • O1, O2direct observations: what was visible or otherwise directly perceived, with location and source/photo references.
  • R1, R2participant-reported information: what someone stated, attributed when known, and marked Reported; not independently verified.
  • I1, I2interpretations: a limited inference from identified observations, explicitly marked as interpretation and never presented as fact or compliance analysis.
  • ISS1, ISS2issues: conditions requiring review, clarification, testing, or action; state urgency only when the source supports it.
  • F1, F2proposed follow-ups: candidate actions, with proposed owner and due date when known; not canonical tasks until selected through /as:tasklist.

Link photographs and files to the exact item they support. Do not describe an image you cannot inspect. Do not use a participant statement to populate Direct Observations, even when the statement seems plausible. A contractor statement that a wall is non-load-bearing remains participant-reported and unverified; it is not a structural finding.

For each inaccessible, obscured, untested, or concealed area relevant to the visit, record a specific limitation. Absence of an observation is not evidence of absence.

Step 4 — Write a collision-safe report

Use:

site-reports/YYYY-MM-DD-<descriptive-slug>.md

Use the visit date in the filename. Resolve and show the exact project-relative target. If it exists:

  • When the user explicitly asked to revise that exact report, preserve the original visit date, Created date, and sources; set Updated to today and add a concise revision note.
  • Otherwise preserve the existing file and select the next available suffix: -02, -03, and so on. Check the suffixed path before writing. Never overwrite merely because date and title match.

Resolve the bundled template relative to the loaded skills/site-visit-report/SKILL.md when the harness exposes that path. On Claude Code, ${CLAUDE_PLUGIN_ROOT}/skills/site-visit-report/templates/site-visit-report.md is the fallback. If bundled resources are unavailable, reproduce every section required by Step 3 plus visit metadata, sources, photos/files, limitations, and the disclaimer behavior below.

Write the complete report and confirm its saved path. At this point stop all other mutation and verify that PROJECT.md, decisions/, and TASKS.md were not changed by saving the report.

Step 5 — Apply the professional boundary

Append the canonical disclaimer block at the very end when either condition is true:

  • the report is client-facing, authority-facing, issued externally, or may reasonably be submitted to a client or authority; or
  • the artifact includes regulated analysis or conclusions involving code, zoning, occupancy, life safety, structural or MEP adequacy, accessibility, environmental risk, or similar professional judgment.

If uncertain whether the report will be submitted externally, include it. A purely internal administrative draft containing observations, logistics, photo references, and follow-up coordination—but no regulated conclusion—may omit it. Omission does not relax the concealed-condition, certainty, or compliance prohibitions.

When the bundled rule is available, read rules/professional-disclaimer.md. Otherwise use this exact fallback as the final content, with one blank line between the block and marker:

> Disclaimer: This is an AI-generated analysis for preliminary planning purposes. All findings must be verified by a licensed professional before use in design, permitting, or regulatory submissions.

<!-- architecture-studio:requires-disclaimer -->

The marker is a terminal sentinel. Do not place anything after it.

Step 6 — Preview optional promotions

After saving, present separate candidate lists:

  • Potential project facts: selected observations that may warrant verification and entry through /as:project remember. Do not propose reported information or interpretation as verified facts.
  • Durable decisions: an explicitly documented choice, if any, suitable for /as:project record-decision; observations and issues are not decisions.
  • Canonical tasks: selected F# follow-ups suitable for /as:tasklist.

Explain that no downstream record has changed. Require exact item labels such as O2, F1, or none. Before handoff, search the owning record for the same project-relative report link and item label; when found, show it and offer link/update/new rather than silently duplicating it. Do not promote unselected siblings.

Step 7 — Hand off without hard dependencies

When direct skill invocation exists, invoke the owning skill once per selected record type and pass the report path and exact labels. The destination skill owns confirmation and mutation.

When direct invocation is unavailable, leave the report intact and print only applicable copyable commands in this form:

/as:project remember selected verified observation O2 from site-reports/2026-07-21-roof-walkthrough.md; preserve the source backlink and verify it before recording as a project fact
/as:project record-decision selected durable choice from site-reports/2026-07-21-roof-walkthrough.md; preserve the exact source item label and backlink
/as:tasklist import selected follow-ups F1, F3 from site-reports/2026-07-21-roof-walkthrough.md; preview duplicates and changes before writing

If nothing is selected, finish with the site report only.

Other skills for the same job

different authors, same section of the catalogue
Startup Analyst
by ComeOnOliver
×2

Expert startup business analyst specializing in market sizing, financial modeling, competitive analysis, and strategic planning for early-stage companies. Use PROACTIVELY when the user asks about market opportunity, TAM/SAM/SOM, financial projections, unit economics, competitive landscape, team planning, startup metrics, or business strategy for pre-seed through Series A startups.

5k tokens
Team Composition Analysis
by ComeOnOliver
×2

This skill should be used when the user asks to "plan team structure", "determine hiring needs", "design org chart", "calculate compensation", "plan equity allocation", or requests organizational design and headcount planning for a startup.

5k tokens
Bulk Rnaseq
by K-Dense-AI
×1

End-to-end bulk RNA-seq orchestrator — takes raw FASTQ reads through QC and trimming (FastQC, fastp/Trim Galore), alignment and quantification (STAR, Salmon, featureCounts), assembles a gene-level counts matrix, then hands off to differential expression (pydeseq2), pathway/GSEA enrichment (pathway-enrichment), and publication figures (scientific-visualization). Use whenever the user has bulk RNA-seq reads or quant output and wants a complete, reproducible differential-expression workflow — e.g. "analyze my RNA-seq", "FASTQ to DESeq2", "run nf-core/rnaseq", "STAR/Salmon quantification", "build a counts matrix for DESeq2", or "go from reads to differentially expressed genes and enriched pathways". Routes between an nf-core/rnaseq (Nextflow) path and a standalone STAR/Salmon path, and covers experimental design, strandedness, and QC gates. For single-cell RNA-seq use the scanpy skill instead.

13k tokens scripts
Generate Status Report
by openai
vendor ×1

Generate project status reports from Jira issues and publish to Confluence. When an agent needs to: (1) Create a status report for a project, (2) Summarize project progress or updates, (3) Generate weekly/daily reports from Jira, (4) Publish status summaries to Confluence, or (5) Analyze project blockers and completion. Queries Jira issues, categorizes by status/priority, and creates formatted reports for delivery managers and executives.

6k tokens scripts
Us Market Bubble Detector
by nicepkg
×1

Evaluates market bubble risk through quantitative data-driven analysis using the revised Minsky/Kindleberger framework v2.1. Prioritizes objective metrics (Put/Call, VIX, margin debt, breadth, IPO data) over subjective impressions. Features strict qualitative adjustment criteria with confirmation bias prevention. Supports practical investment decisions with mandatory data collection and mechanical scoring. Use when user asks about bubble risk, valuation concerns, or profit-taking timing.

22k tokens scripts
Gws Workflow Standup Report
by googleworkspace
vendor

Google Workflow: Today's meetings + open tasks as a standup summary.

278 tokens
Recipe Create Events From Sheet
by googleworkspace
vendor

Read event data from a Google Sheets spreadsheet and create Google Calendar entries for each row.

223 tokens
Baoyu Diagram
by JimLiu

Create professional, dark-themed SVG diagrams of any type — architecture diagrams, flowcharts, sequence diagrams, structural diagrams, mind maps, timelines, illustrative/conceptual diagrams, and more. Use this skill whenever the user asks for any kind of technical or conceptual diagram, visualization of a system, process flow, data flow, component relationship, network topology, decision tree, org chart, state machine, or any visual representation of structure/logic/process. Also trigger when the user says "画个图" "画一个架构图" "diagram" "flowchart" "sequence diagram" "draw me a ..." or uploads content and asks to visualize it. Output is always a standalone .svg file.

7k tokens scripts

How to use it

Copy the folder

Take alpacalabsllc/site-visit-report from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.