Install
$ agentstack add skill-lishix520-jtbd-skills-jtbd-context-explorer ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README — it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
JTBD Context Explorer
Extract and organize evidence about the circumstances in which a person seeks to make progress. Preserve the difference between direct evidence, inference, and unresolved uncertainty.
This skill investigates context around a possible job. It does not define or validate a Core Functional Job, map a workflow, generate desired outcomes, compute market opportunity, or select a strategy.
Required Input
Accept one or more of:
- Interview excerpts or transcripts
- Customer feedback, reviews, support tickets, or open-text survey responses
- A scenario description with clearly identified source and limitations
If no source material is supplied, return analysis_status: insufficient_input. Do not invent a customer circumstance, motivation, competing alternative, or non-consumption behavior.
Scope
Extract evidence for:
- Circumstance: A concrete situation, trigger, constraint, event, or change that creates pressure to make progress.
- Desired Progress: A stated or strongly implied change from a current situation toward a better future situation.
- Current Approach: What the person currently does, including a product, manual process, workaround, delay, delegation, avoidance, or doing nothing.
- Switching Trigger: A reported event or threshold that made the current approach inadequate.
- Non-Consumption: Evidence that the person does not use an available solution because it is inaccessible, unaffordable, too complex, unsuitable, unavailable, prohibited by policy, security-restricted, or not worth adopting.
- Feature Requests: Statements proposing specific product capabilities, buttons, or tools.
- Constraints: Time, money, policy, environment, access, skill, compatibility, privacy, safety, or organizational restrictions.
- Competing Alternatives: Solutions, behaviors, workarounds, or "do nothing" approaches used to make progress.
- Emotional Signal: Evidence of a desired feeling or avoided feeling.
- Social Signal: Evidence about how the person wants to be perceived by another person or group.
Evidence Rules
- Preserve direct statements as excerpts with source IDs.
- Label every interpretation as
inferred; never present it as direct fact. - Strongly implied desired progress is ALWAYS
status: inferred. Only explicit customer statements of intent aredirect_evidence. - Do not infer causal relationships when the source only shows correlation.
- Record proposed solution statements under
feature_requests; do not convert a feature request into a validated Core Functional Job, current approach, or competing alternative without independent evidence. - Do not treat a complaint as proof of a broad market pattern or non-consumption.
- Do not classify something as non-consumption unless the material shows absence, avoidance, inability, or refusal to use a relevant alternative.
- Treat emotional and social language as separate signals, not as Core Functional Jobs or ODI desired outcomes.
- Record contradictions rather than resolving them through guesswork.
- Return the smallest next research question that would reduce the most consequential uncertainty.
Procedure
- Assign stable source IDs to supplied material.
- Extract direct excerpts without paraphrasing their evidentiary meaning.
- Classify each excerpt into one or more evidence categories.
- Record proposed solution statements under
feature_requests. - Build a circumstance record only when source material identifies a concrete situation, trigger, constraint, or change.
- Identify current approaches and alternatives, including delay and doing nothing.
- Extract top-level constraints (policy, security, time, money, etc.).
- Mark non-consumption only when the required evidence threshold is met.
- Separate emotional and social signals from functional progress.
- List contradictions, unsupported hypotheses, and evidence gaps.
- Produce research questions; do not turn gaps into conclusions.
Output Format
analysis_status: evidence_extracted | insufficient_input
sources:
- id: ""
type: interview | review | feedback | support_ticket | scenario
limitations: []
circumstances:
- id: ""
statement: ""
status: direct_evidence | inferred
evidence:
- source_id: ""
excerpt: ""
assumptions: []
constraint_ids: []
desired_progress:
- id: ""
statement: ""
status: direct_evidence | inferred
evidence: []
assumptions: []
current_approaches:
- id: ""
statement: ""
type: product | manual_process | workaround | delegation | delay | avoidance | do_nothing | unknown
status: direct_evidence | inferred
evidence: []
assumptions: []
switching_triggers:
- id: ""
statement: ""
status: direct_evidence | inferred
evidence: []
assumptions: []
non_consumption:
- id: ""
statement: ""
barrier_type: access | affordability | complexity | suitability | availability | policy_or_regulation | security_or_privacy | perceived_value | unknown
status: direct_evidence | inferred
evidence: []
assumptions: []
feature_requests:
- id: ""
statement: ""
status: direct_evidence | inferred
evidence: []
assumptions: []
constraints:
- id: ""
statement: ""
type: time | money | policy_or_regulation | security_or_privacy | environment | access | skill | compatibility | privacy | safety | organizational | unknown
status: direct_evidence | inferred
evidence: []
assumptions: []
competing_alternatives:
- id: ""
statement: ""
type: product | manual_process | workaround | delegation | delay | avoidance | do_nothing | unknown
status: direct_evidence | inferred
evidence: []
assumptions: []
emotional_signals:
- id: ""
statement: ""
status: direct_evidence | inferred
evidence: []
assumptions: []
social_signals:
- id: ""
statement: ""
audience: ""
status: direct_evidence | inferred
evidence: []
assumptions: []
contradictions: []
evidence_gaps: []
next_research_question: ""
Reference
Read references/context-exploration-rules.md before:
- Categorizing proposed solutions into
feature_requests - Deciding whether a phrase proves non-consumption vs doing nothing
- Determining whether desired progress is
direct_evidenceorinferred - Separating an emotional signal from a social signal
- Extracting top-level constraints
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: lishix520
- Source: lishix520/jtbd-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.