Call Arithmetic Accuracy Auditor
Offline experimental transcript auditor that recomputes the arithmetic a CALL-E voice agent SPOKE (quantity-times-price totals, line-item sums, percent-off results, balance-due subtraction, clock durations) against the agent's own stated factors and grades ALL_COMPUTATIONS_CORRECT, ARITHMETIC_ERRORS_FOUND, UNVERIFIABLE_COMPUTATIONS_ONLY, or NO_SPOKEN_COMPUTATIONS, plus a factors-first goal crafter; not a pricing database, a billing system, or authorization to act.
Install
npx skills add CALLE-AI/awesome-phone-call-agents --skill call-arithmetic-accuracy-auditorgit clone --depth 1 https://github.com/CALLE-AI/awesome-phone-call-agents.git /tmp/awesome-phone-call-agents
mkdir -p ~/.claude/skills
cp -r /tmp/awesome-phone-call-agents/skills/call-arithmetic-accuracy-auditor ~/.claude/skills/call-arithmetic-accuracy-auditorPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# call-arithmetic-accuracy-auditor
> **A voice agent that states its factors out loud and then speaks a
> wrong total has committed the error in front of a customer. A human
> skimming the transcript hears a confident number - the audit has to
> redo the multiplication to catch it.**
Most QA measures whether a total was disclosed. Almost none measures
whether the total was arithmetically true given the numbers the agent
itself just spoke. This skill reads a finished call transcript,
recomputes every spoken computation from the agent's own stated factors,
and grades each one CORRECT, INCORRECT, or UNVERIFIABLE - then rolls the
sentence grades into one verdict a reviewer can act on.
## Why This Skill Exists
LLM arithmetic is not reliable enough to skip verification. Dziri et al.
("Faith and Fate", arXiv:2305.18654) show transformers solve
composition problems by pattern matching, not computation. Mirzadeh et
al. (GSM-Symbolic, arXiv:2410.05229) show accuracy degrades when only
the numbers change, so a clean-sounding factor sentence does not
guarantee a correct total. Wei et al. (arXiv:2505.15000) show spoken
math is worse still for cascade ASR+LLM pipelines - exactly CALL-E's
architecture. And the FTC's Policy Statement on Deception makes a wrong
spoken total a material misrepresentation to a consumer acting
reasonably. Verification therefore has to happen outside the model,
which is what this auditor is.
## When To UseDetails
- Skill name
call-arithmetic-accuracy-auditor- Repository
- CALLE-AI/awesome-phone-call-agents
- Path
skills/call-arithmetic-accuracy-auditor/SKILL.md- License
- MIT
- Repo updated
- 14 November 2026
FAQ
What does the Call Arithmetic Accuracy Auditor skill do?
Offline experimental transcript auditor that recomputes the arithmetic a CALL-E voice agent SPOKE (quantity-times-price totals, line-item sums, percent-off results, balance-due subtraction, clock durations) against the agent's own stated factors and grades ALL_COMPUTATIONS_CORRECT, ARITHMETIC_ERRORS_FOUND, UNVERIFIABLE_COMPUTATIONS_ONLY, or NO_SPOKEN_COMPUTATIONS, plus a factors-first goal crafter; not a pricing database, a billing system, or authorization to act.
How do I install the Call Arithmetic Accuracy Auditor skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add CALLE-AI/awesome-phone-call-agents --skill call-arithmetic-accuracy-auditor. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Call Arithmetic Accuracy Auditor popular?
It lives in CALLE-AI/awesome-phone-call-agents, which has 116 GitHub stars and 512 forks.
Related skills
Accesscall
116CALLE-AI/awesome-phone-call-agents
Conduct phone-based accessibility intake interviews for users who cannot complete web-based accessibility audit forms (screen reader fatigue, motor impairment, low vision, cognitive load), and produce a structured result mapped to VPAT 2.4 / Section 508 conformance reporting fields.
Accessible Outing Verifier
116CALLE-AI/awesome-phone-call-agents
Demonstrates advisory accessibility-planning checks with offline fixtures and a proposed bounded CALL-E workflow; use for exploring unknown or qualified venue claims without making calls.
Adherence Memory Callback
116CALLE-AI/awesome-phone-call-agents
Run a consent-based outbound CALL-E medication-adherence phone check-in that remembers each caller across calls, learns side-effect patterns across many callers behind a corroboration gate, and honors "call me back later" by opening the next call with that context.
Call Cancellation Respect Auditor
116CALLE-AI/awesome-phone-call-agents
Offline experimental transcript auditor that detects cancellation requests in CALL-E call transcripts and grades the agent response CANCELLATION_HONORED, SOFT_RETENTION, RETENTION_DARK_PATTERN, CANCELLATION_IGNORED, or REQUEST_UNRESOLVED, plus a cancellation-respect goal crafter; not a legal determination, proof of a violation, or authorization to act.
Call Answer Evasion Auditor
116CALLE-AI/awesome-phone-call-agents
Offline experimental CALL-E transcript helper that grades agent answers to callee direct questions (identity, yes-no, information) as CLEAR, PARTIALLY_CLEAR or EVASIVE with dodge mechanisms, and crafts answer-first goal templates. It is not intent detection, proof of deceit, or authorization to act.
Call Closing Sequence Auditor
116CALLE-AI/awesome-phone-call-agents
Offline heuristic CALL-E transcript skill that audits the call's closing window for conversation-analytic completeness - pre-closing summary, arrangement for next steps, completed terminal exchange, dangling callee questions, business after the farewell, abrupt ends - and emits a clean-closing goal template. It is not a satisfaction measure, proof the call failed, or authorization to act automatically.
Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.